JP2016042321A - Stereo pair image data searching method and search system using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to search for stereo pair image data to be used for a three-dimensional terrain model without using satellite orbit data.SOLUTION: A stereo pair image data searching system includes: a database including a meta data storage part that stores meta data of geographic conditions related to geographical features where image data is photographed, and a published value storage part that stores published values related to photography conditions; a user-side apparatus that includes an input unit used in specifying at least geographic conditions and photography conditions; and a search unit including a search execution part that searches for meta data satisfying the geographic conditions specified by the user-side apparatus in the meta data storage part and outputs identification information included in the meta data as candidate data, and a narrowing part that narrows down the candidate data with the photography conditions and outputs the narrowed data as a search result.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a stereo pair image data search method and a search system using the same.

  Today, commercial satellite image data with high terrestrial resolution is available at an affordable price. Therefore, there is an increasing demand for creating a three-dimensional terrain model using such satellite image data. For example, there is a stereo matching method for creating a three-dimensional terrain model. In this stereo matching method, two satellite image data (stereo pair image data) having different viewpoints are used.

  For example, Japanese Patent Application Laid-Open No. 2012-234374 discloses a stereo pair search device including a satellite image search unit, a satellite position specifying unit, and a stereo pair image extraction unit.

  The satellite image search unit overlaps the image search location based on the geographical coordinate information of the image search location and the polygon information of the satellite image stored in the satellite image capturing database storing the satellite image captured by the artificial satellite. A satellite image having a region is searched.

  At this time, when there are a plurality of searched satellite images, the satellite position specifying unit specifies the position coordinates of the artificial satellite and the center coordinates of the satellite images when the satellite images are captured in three dimensions.

  Then, the stereo pair image extraction unit performs stereo based on the satellite position coordinates and the center coordinates corresponding to the first satellite image and the second satellite image that are an arbitrary pair of the plurality of satellite images. A pair of satellite images satisfying the pair condition is extracted.

JP 2012-234374 A

  However, when searching for a stereo pair image from a satellite image, information on the satellite orbit is required, but information on the satellite orbit is not given to normal commercial satellite image data. For this reason, in the configuration according to Japanese Patent Application Laid-Open No. 2012-234374, it is necessary to provide a satellite position specifying unit to specify a stereo pair image used for a three-dimensional terrain model.

  That is, the satellite position specifying unit, when there are a plurality of searched satellite image data, for each combination of the plurality of satellite image data, the position coordinates of the artificial satellite and the satellite at the time of capturing each satellite image data A process of specifying the center coordinates of the image data in three dimensions is performed.

  At this time, if the number of searched satellite image data increases, the number of combinations of a plurality of satellite image data also becomes enormous, and it is substantially difficult to appropriately search for stereo pair image data used for the three-dimensional terrain model. There is a problem.

  Therefore, a main object of the present invention is to provide a stereo pair image data search method and a search system using the same, which can search stereo pair image data used for a three-dimensional terrain model without using satellite orbit data.

  In order to solve the above problem, an invention relating to a stereo pair image data retrieval system that retrieves a pair of image data used for creating a three-dimensional terrain model as stereo pair image data is provided. A metadata storage unit that stores data, a database that includes a published value storage unit that stores published values related to shooting conditions, and a user-side device that includes at least an input unit used when specifying geographical conditions and shooting conditions A search unit that searches the metadata storage unit for metadata that satisfies the geographical conditions specified by the user side device, and outputs the identification information included in the metadata as candidate data; A search unit including a narrowing-down unit for narrowing down and outputting as a search result. .

  In addition, the invention according to the stereo pair image data search method for searching a pair of image data used for creating a three-dimensional terrain model as stereo pair image data is a geographical condition related to geography in which image data is captured in a metadata storage unit of a database. The stored metadata, the procedure for storing the published values for the shooting conditions in the published value storage unit, the procedure for specifying the geographical conditions and shooting conditions by the user side device, and the specification from the user side device for the search execution unit Search the metadata storage unit for metadata that satisfies the specified geographical conditions, output the identification information contained in this metadata as candidate data, and narrow down the candidate data according to the shooting conditions and output it as a search result And a procedure for causing the above to occur.

  According to the present invention, candidate data is narrowed down according to designated conditions, so that stereo pair image data used for a three-dimensional terrain model can be appropriately searched without using satellite orbit data.

It is a block diagram of the stereo pair image data search system for three-dimensional terrain model creation concerning a 1st embodiment. It is a figure which defines an azimuth and an elevation angle. It is a figure explaining a convergence angle. It is a figure explaining the relationship between the magnitude | size of a convergence angle and the height error which arises in a three-dimensional terrain model, (a) shows the case where a convergence angle is small, (b) shows the case where a convergence angle is large. It is a figure explaining the relationship between ground resolution and the height error which arises in a three-dimensional terrain model, (a) shows the case where ground resolution is large, (b) shows the case where ground resolution is small. It is a figure explaining the height error of the three-dimensional terrain model with respect to CE90, (a) shows the case where CE90 is large, and (b) shows the case where CE90 is small. It is a flowchart which shows the stereo pair image data search method. It is a block diagram of the stereo pair image data search system for three-dimensional terrain model creation concerning 2nd Embodiment. It is a flowchart which shows the stereo pair image data search method.

<First Embodiment>
A first embodiment of the present invention will be described. FIG. 1 is a block diagram of a stereo pair image data search system 2A for creating a three-dimensional terrain model according to the present embodiment. The stereo pair image data search system 2 </ b> A includes a search unit 10, a user side device 20, and a database 30.

  The database 30 includes a metadata storage unit 31 that stores metadata related to the geographical conditions of the satellite image data, and a published value storage unit 32 that stores published values related to imaging conditions such as ground resolution and horizontal position accuracy index. .

  The metadata storage unit 31 includes a shooting range storage unit 31a that stores geographic information (position information) of satellite image data such as latitude and longitude and UTM (Universal Transverse Mercator) coordinates, and shooting angle information of azimuth and elevation angles of satellite image data. Includes an azimuth / elevation angle storage unit 31b.

  The published value storage unit 32 includes a ground resolution storage unit 32a that stores the ground resolution of satellite image data, and a horizontal position accuracy index storage unit 32b that stores horizontal position accuracy indexes in satellite image data such as CE90.

  It does not matter whether the satellite image data is commercial satellite image data such as IKONOS or QuickBird. Each piece of information stored in the metadata storage unit 31 or the published value storage unit 32 is associated with one satellite image data by identification information (ID).

  The user side device 20 includes an input unit 21 for inputting various information used for searching for stereo pair image data suitable for creating a three-dimensional terrain model, and an output unit 22 for displaying information of the searched stereo pair image data. Prepare.

  The input unit 21 specifies a geographical condition including a keyboard and the like used for inputting information (hereinafter referred to as geographical condition information G1) regarding an area (a geographical position where a three-dimensional terrain model is created) that the user wants to search for stereo pair image data. A photographing condition designating part 21b for setting information on the part 21a, the angle of convergence, the ground resolution, and the horizontal position accuracy index as a search parameter threshold value.

  Hereinafter, information related to the convergence angle, the ground resolution, and the horizontal position accuracy index is collectively referred to as imaging condition information G2. Also, the search parameter threshold for the convergence angle is described as the convergence angle threshold, the search parameter threshold for the ground resolution is described as the ground resolution threshold, and the search parameter threshold for the horizontal position accuracy index is described as the horizontal position accuracy index threshold. That is, the imaging condition information G2 includes a convergence angle threshold value, a ground resolution threshold value, and a horizontal position accuracy index threshold value.

  The search unit 10 searches the metadata of the stereo pair image data from the shooting range storage unit 31a based on the geographical condition information G1, and outputs this as candidate data. The search execution unit 11 outputs candidate data based on the shooting condition information G2. The narrowing down part 12 is provided.

  The narrowing-down unit 12 is a convergence angle narrowing device 12a that narrows down candidate data based on the convergence angle threshold value of the shooting condition information G2, a resolution narrowing device 12b that narrows down candidate data based on the ground resolution threshold value of the shooting condition information G2, and shooting condition information. A position accuracy narrower 12c that narrows down candidate data based on the horizontal position accuracy index threshold of G2 is included.

  Next, (1) the convergence angle of stereo pair image data, (2) the ground resolution of satellite image data, and (3) the horizontal position accuracy index of satellite image data will be described.

(1) Convergence Angle of Stereo Pair Image Data The convergence angle is calculated according to Equation 1 described later using the azimuth angle and elevation angle in the direction from the observation point toward the satellite (line of sight) included in the metadata of the satellite image data. .

  FIG. 2 is a diagram for defining the azimuth angle α and the elevation angle β. Now, a line connecting the observation point and the satellite is defined as an imaging direction line K1, and a line obtained by projecting the imaging method ray K1 on the ground (a plane including the north direction and the east direction) is defined as a projection line K2. At this time, the azimuth angle α is an angle (eastward) formed by the projection line K2 and a north direction line (hereinafter, north direction line) K3. The elevation angle β is an angle formed by the shooting direction line K1 and the projection line K2. These azimuth angle α and elevation angle β are included in the metadata of the satellite image data.

  When creating a 3D terrain model by using the stereo matching method using satellite image data, a pair of satellite image data (stereo pair image data) is required to capture the same region from multiple directions (photographed from different directions of the line of sight). It becomes. At this time, the angle (convergence angle) formed by each imaging method light beam of the two satellite image data becomes important.

  FIG. 3 is a diagram for explaining the convergence angle. Here, the observation points are photographed by the first satellite and the second satellite, the first satellite image is the satellite image 1, the second satellite image is the satellite image 2, and the satellite image 1 and the satellite image 2 are stereo pair images. Suppose it is used as data. The line toward the first satellite is the imaging direction line K1_a, the line toward the second satellite is the imaging direction line K1_b, the azimuth angle of the first satellite is the azimuth angle α_a, the elevation angle is the elevation angle β_a, and the azimuth angle of the second satellite is the azimuth direction. Assume that the angle α_b and the elevation angle are the elevation angle β_b.

At this time, the convergence angle γ of the first satellite and the second satellite is
γ = cos ^-1 (cosβ_a * sinα_a * cosβ_b * sinα_b +
cosβ_a * cosα_a * cosβ_b * cosα_b +
sinβ_a * sinβ_b)… (1)
Is given by Equation 1.

  The height error that occurs in the three-dimensional terrain model depends on the size of the convergence angle γ. FIG. 4 is a diagram for explaining the relationship between the size of the convergence angle γ and the height error generated in the three-dimensional terrain model. FIG. 4A shows a case where the convergence angle is small, and FIG. The large case is shown.

  When a pixel shift occurs during stereo matching of a pair of satellite image data, an error D occurs in the height generated in the three-dimensional terrain model. The error D increases as the convergence angle γ decreases (see FIGS. 4A and 4B).

  Also, when the convergence angle γ is large, a difference in satellite images is likely to occur due to a difference in the direction of collapse of a building or a tree. This difference causes a matching error in the stereo matching process.

  Therefore, in this embodiment, when searching for stereo pair image data, the range of the convergence angle γ can be specified so that the error of the created three-dimensional terrain model becomes small. Specifically, the convergence angle threshold value can be set to the imaging condition specified by the imaging condition specifying unit 21b. The convergence angle threshold means a range of convergence angles allowed as stereo pair image data.

(2) Ground resolution of satellite image data The ground resolution is defined as an actual distance indicated by one pixel in the satellite image data. Therefore, the error of the three-dimensional terrain model created by the stereo matching method changes according to the size of the ground resolution. 5A and 5B are diagrams for explaining the relationship between the ground resolution and the height error generated in the three-dimensional terrain model. FIG. 5A shows a case where the ground resolution is large, and FIG. 5B shows a case where the ground resolution is small. ing.

  When a one-pixel shift occurs due to a stereo matching error, the height error of the three-dimensional terrain model is small when the ground resolution of the satellite image data is small. That is, the stereo pair image data used for creating the three-dimensional terrain model has a smaller height error when the ground resolution is smaller.

  Therefore, in the present embodiment, the ground resolution threshold value can be set in the shooting condition information specified by the shooting condition specifying unit 21b. The ground resolution threshold means a range of ground resolution allowed as stereo pair image data.

(3) Horizontal position accuracy index of satellite image data The horizontal position accuracy index can be defined as an index indicating the accuracy (coincidence) between the position information of the satellite image data and the actual position information. As an example of the horizontal position accuracy index, there is a CE 90 that samples the errors between the horizontal position information of the satellite image data and the actual horizontal position information at a plurality of points and uses the error value of the top 90% of the total value as an index.

  6A and 6B are diagrams for explaining the height error of the three-dimensional terrain model with respect to the CE 90. FIG. 6A shows a case where the CE 90 is large and FIG. 6B shows a case where the CE 90 is small. When a three-dimensional terrain model is created by the stereo matching method, when the horizontal position accuracy index is poor (for example, the value of CE90 is large), the height error of the three-dimensional terrain model becomes large as shown in FIG. Therefore, it is preferable to use satellite image data having a small horizontal position accuracy index for creating the three-dimensional terrain model.

  Therefore, in the present embodiment, the horizontal position accuracy index threshold value can be set in the shooting condition information specified from the shooting condition specifying unit 21b. The horizontal position accuracy index threshold means a range of horizontal position accuracy index allowed as stereo pair image data.

  Next, a stereo pair image data search method in the stereo pair image data search system 2 having such a configuration will be described. FIG. 7 is a flowchart showing a stereo pair image data search method. The stereo pair image data search method is roughly divided into candidate data search processing (step SA1 to step SA3), convergence angle narrowing processing (step SA4 to step SA6), resolution narrowing processing (step SA7 to step SA9), This is divided into position accuracy narrowing-down processing (steps SA10 to SA11).

Step SA1: (Candidate data search process)
First, the geographical condition specifying unit 21a of the user device 20 acquires the geographical condition information G1. This geographical condition information G1 is information related to a geographical area when searching for stereo pair image data, and is set by the user from the geographical condition designating unit 21a. The geographic condition designating unit 21a outputs the input geographic condition information G1 to the search execution unit 11.

  Steps SA2 and SA3: The search execution unit 11 searches the shooting range storage unit 31a based on the geographical condition information G1. By this search, the metadata of the satellite image at the geographical position satisfying the geographical condition information G1 is acquired. Since the metadata is associated with the original satellite image data as identification information (ID), the data acquired by the search execution unit 11 is identification information. At this time, there may be a plurality of metadata that satisfy the geographical condition. In this case, a plurality of pieces of identification information are acquired. Therefore, data including a plurality of acquired identification information is described as candidate data. The search execution unit 11 determines whether or not the number of pieces of identification information included in the acquired candidate data is one or less. If the number is one or less, the search execution unit 11 proceeds to step SA12 to indicate that there is no candidate data. The result is output, and if there are two or more, the process proceeds to step SA4.

Steps SA4 to SA6: (Convergence angle narrowing process)
The search execution unit 11 outputs the candidate data to the convergence angle narrower 12a. Imaging condition information G2 is input to the convergence angle narrowing device 12a from the imaging condition designating unit 21b. Therefore, the convergence angle narrowing down device 12a extracts a convergence angle threshold value related to the convergence angle from the imaging condition information G2.

  Then, the convergence angle narrower 12a searches the azimuth / elevation angle storage unit 31b using the identification information included in the candidate data as a search keyword, and acquires the azimuth and elevation angles of the satellite image that matches the identification information. Therefore, the set data of the azimuth angle and the elevation angle is acquired by the number of pieces of identification information included in the candidate data from the search execution unit 11. If a plurality of pieces of identification information exist in candidate data, stereo image data pairs can be formed by the number of combinations. Therefore, the convergence angle narrowing device 12a calculates the convergence angle by applying the azimuth angle and elevation angle of the identification information forming a pair to Equation 1. Then, it is determined whether the calculated convergence angle satisfies the extracted convergence angle threshold value. One pair of identification information corresponding to the convergence angle satisfying the convergence angle threshold is the data content of the convergence angle narrowing result data.

  Such a convergence angle narrowing process is performed on all pairs of a plurality of identification information, and the result is output as convergence angle narrowing result data (candidate data). Accordingly, the convergence angle narrowing result data (candidate data) includes a plurality of pairs of identification information.

  Next, the convergence angle narrowing device 12a determines the output destination of the convergence angle narrowing result data. That is, when the convergence angle narrowing result data includes one or more pairs of identification information, the convergence angle narrowing result data is output to the resolution narrowing device 12b, and the process proceeds to step SA7. On the other hand, if there is no identification information included in the convergence angle narrowing result data, the process proceeds to step SA12 as a search result and outputs to the output unit 22 that there is no convergence angle narrowing result data.

Steps SA7 to SA9: (resolution narrowing process)
The resolution narrowing device 12b receives the convergence angle narrowing result data (candidate data) from the convergence angle narrowing device 12a and the photographing condition information G2 from the photographing condition designating unit 21b. Therefore, the resolution narrower 12b extracts a ground resolution threshold value from the imaging condition information G2. Then, the ground resolution storage unit 32a is searched using the identification information included in the convergence angle narrowing result data as a keyword, and the ground resolution of the satellite image associated with the identification information is acquired.

  Thereafter, the resolution narrower 12b determines whether or not the acquired ground resolution satisfies the ground resolution threshold value. That is, the convergence angle narrowing result data (candidate data) includes a plurality of pairs of identification information. Therefore, the resolution narrower 12b acquires the ground resolution of the satellite image corresponding to the identification information using each identification information as a keyword. Then, it is determined whether the acquired ground resolution satisfies a ground resolution threshold value. When the ground resolutions of the pair corresponding to the paired identification information both satisfy the ground resolution threshold, the result is included in the resolution narrowing result data (candidate data).

  Next, when the resolution narrowing result data includes one or more pairs of identification information, the resolution narrowing down device 12b outputs the resolution narrowing result data to the position accuracy narrowing down device 12c and proceeds to Step SA10. On the other hand, when there is no identification information included in the resolution narrowing result data, the fact that there is no resolution narrowing result data goes to step SA12 as a search result and is output to the output unit 22.

Step SA10: (Position accuracy narrowing process)
The position accuracy narrowing device 12c receives resolution narrowing result data from the resolution narrowing device 12b, and also receives photographing condition information G2 from the photographing condition designating unit 21b. Therefore, the position accuracy narrower 12c extracts a horizontal position accuracy index threshold value from the imaging condition information G2. Further, the horizontal position accuracy index storage unit 32b is searched using the identification information included in the resolution narrowing result data as a keyword, and the horizontal position accuracy index of the satellite image associated with the matching identification information is acquired. Since the resolution narrowing result data includes a plurality of pairs of identification information, the number of horizontal position accuracy indices acquired is twice the number of pairs.

  Then, the position accuracy narrower 12c determines whether each horizontal position accuracy index satisfies a horizontal position accuracy index threshold. When both the horizontal position accuracy indices corresponding to the paired identification information satisfy the horizontal position accuracy index threshold, the pair identification information is included in the position accuracy narrowing result data. If either one does not satisfy the horizontal position accuracy index threshold, the result is that there is no position accuracy narrowing result data. The result of performing the position accuracy narrowing process for the identification information of all pairs included in the resolution narrowing result data in this way is set as position accuracy narrowing result data (candidate data), and the process proceeds to step SA12 to output the output unit 22 Is output.

  Stereo pair image data suitable for creating a three-dimensional terrain model using the search results obtained through such a procedure (the search results include one or more pairs of identification information corresponding to the satellite image data). Can be acquired. That is, it is possible to create a high-quality three-dimensional terrain model with less noise caused by the stereo matching method and less noise.

Second Embodiment
Next, a first embodiment of the present invention will be described. In addition, about the same structure as 1st Embodiment, description is abbreviate | omitted suitably using the same code | symbol.

  The metadata of satellite image data may include observation date and time and cloud coverage. Therefore, in the present embodiment, by using information on observation date and time and cloud coverage, it is possible to search for stereo pair image data more suitable for creating a three-dimensional terrain model by the stereo matching method.

  FIG. 8 is a block diagram of a stereo pair image data search system 2B for creating a three-dimensional terrain model according to the present embodiment.

  The influence on the three-dimensional terrain model due to the difference in the observation date and time of the stereo pair image will be described. In stereo pair image data with different observation dates and times, for example, there may be a structure in one satellite image and no structure in the other satellite image. Further, there is a case where one satellite image has a cloud and the other satellite image has no cloud (including a case where the amount of clouds is different). When such a stereo pair image data is used to create a three-dimensional terrain model, there is a disadvantage that noise is generated depending on the presence or absence of a structure.

  Therefore, in the present embodiment, threshold values are set for differences in observation date / time and the presence / absence of clouds, and candidate data is narrowed down by this threshold value. For this reason, the metadata storage unit 31 of the database 30 is additionally provided with a cloud coverage storage unit 31c that stores metadata related to the cloud coverage, and an observation date storage unit 31d that stores metadata related to the observation date and time. Further, the narrowing unit 12 of the search unit 10 is additionally provided with a cloud coverage narrowing device 12d and an observation date narrowing device 12e.

  The shooting condition information specified by the shooting condition specifying unit 21b includes a cloud coverage threshold value relating to the cloud coverage rate and an observation date threshold value relating to the observation date and time.

  FIG. 9 is a flowchart showing the stereo pair image data search method according to this embodiment. The flow chart is generally the same as the flowchart showing the stereo pair image data search method in the first embodiment shown in FIG. That is, steps SB1 to SB3 are candidate data search processes corresponding to steps SA1 to SA3, steps SB7 to SB9 are convergence angle narrowing processes corresponding to steps SA4 to SA6, and steps SB13 to SB18 are resolutions corresponding to steps SA7 to SA12. A narrowing process and a position accuracy narrowing process.

  Steps SB4 to SB6 are added for the cloud coverage rate narrowing process based on the cloud coverage rate, and steps SB10 to SB12 are added for the observation date narrowing process based on the observation date and time. Hereinafter, description of the same processing is omitted, and only the added processing is described.

Steps SB4 to SB6: (Cloud coverage narrowing process)
The cloud coverage narrowing device 12d extracts a cloud coverage threshold value related to the cloud coverage from the photographing condition information G2. The cloud coverage narrowing device 12d searches the cloud coverage storage unit 31c using the identification information included in the candidate data from the search execution unit 11 as a keyword, and cloud cover of the satellite image data corresponding to the identification information. Get rate.

  Then, the cloud coverage narrowing device 12d determines whether or not the acquired cloud coverage satisfies the cloud coverage threshold. Since the candidate data includes a plurality of pieces of identification information, the cloud coverage narrowing process is performed for each piece of identification information. The identification information that satisfies the cloud coverage threshold is output as cloud coverage narrowing result data (candidate data).

  At this time, if the cloud cover narrowing result data (candidate data) contains one or less identification information, the process proceeds to step SB18, and the fact that there is no candidate data is output to the output unit 22 as a search result. On the other hand, when two or more pieces of identification information are included in the cloud cover narrowing result data (candidate data), the process proceeds to step SB7 and the convergence angle narrowing process is performed.

Steps SB10 to SB12: (observation date narrowing process)
The observation date / time narrowing down device 12e extracts an observation date / time threshold related to the observation date / time from the imaging condition information G2. The observation date / time narrowing down device 12e searches the observation date / time storage unit 31d using the identification information included in the convergence angle narrowing result data from the convergence angle narrowing down device 12a as a keyword, and the satellite image corresponding to the identification information is searched. Get the observation date.

  Then, the observation date narrower 12e narrows down the convergence angle narrowing result data by determining whether or not the acquired observation date meets the observation date threshold. At this time, since the convergence angle narrowing-down result data includes a plurality of pairs of identification information, the number of observation dates and times acquired is twice the number of pairs. Therefore, when the observation date and time corresponding to the paired identification information satisfy both the observation date and time thresholds, the observation date and time narrowing down device 12e sets the pair identification information as observation date and time narrowing result data (candidate data).

  If one or more pairs of identification information is included in the observation date / time narrowing result data (candidate data), the process proceeds to step SB13, and the resolution narrowing process is performed. If there is no identification information, the process proceeds to step SB18. The fact that there is no search result is output to the output unit 22.

  By performing such processing, it is possible to narrow down the observation date and time, the presence or absence of clouds, and the like, and it becomes possible to search for stereo pair image data for creating a three-dimensional terrain model with suppressed noise.

  A part or all of the above embodiment can be described as in the following supplementary notes, but is not limited thereto.

(Appendix 1)
A stereo pair image data search system for searching a pair of image data used for creating a three-dimensional terrain model as stereo pair image data,
A metadata storage unit that stores metadata of geographic conditions related to the geography in which the image data was captured, and a database that includes a published value storage unit that stores published values related to the shooting conditions;
A user side device including at least an input unit used for designating the geographical condition and the photographing condition;
A search execution unit that searches the metadata storage unit for the metadata that satisfies the geographic condition specified by the user side device, and outputs identification information included in the metadata as candidate data, and the imaging condition A search unit including a narrowing unit that narrows down the candidate data and outputs the result as a search result;
A stereo pair image data retrieval system comprising:
(Appendix 2)
The stereo pair image data search system according to appendix 1,
The imaging condition includes a cloud coverage threshold value that defines a cloud coverage rate when the image data is captured,
The metadata storage unit includes a cloud coverage storage unit that stores the cloud coverage of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the cloud coverage storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and adds the acquired metadata to the acquired metadata. A cloud coverage refiner is included that narrows down the candidate data by comparing the cloud coverage and the cloud coverage threshold included.
A stereo pair image data retrieval system characterized by that.
(Appendix 3)
The stereo pair image data search system according to appendix 1 or 2,
The shooting conditions include a convergence angle threshold that defines a convergence angle of the stereo pair image data used when creating the three-dimensional terrain model,
The metadata storage unit includes an azimuth / elevation angle storage unit that stores the metadata of the azimuth and elevation when the image data is captured in association with the identification information,
The narrowing-down unit searches the azimuth / elevation angle storage unit by using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the metadata A convergence angle narrowing device that calculates the convergence angle using the azimuth angle and elevation angle, and narrows down the candidate data by comparing the calculated convergence angle and the convergence angle threshold,
A stereo pair image data retrieval system characterized by that.
(Appendix 4)
The stereo pair image data search system according to any one of appendices 1 to 3,
The shooting conditions include an observation date threshold that defines the observation date and time when the image data was shot,
The metadata storage unit includes an observation date storage unit that stores the observation date and time of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the observation date storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the acquired metadata Including an observation date and time narrower that narrows down the candidate data by comparing the observation date and time and the observation date and time threshold.
A stereo pair image data retrieval system characterized by that.
(Appendix 5)
The stereo pair image data search system according to any one of appendices 1 to 4,
The shooting conditions include a ground resolution threshold value that defines the ground resolution when shooting the image data,
The metadata storage unit includes the ground resolution storage unit that stores the ground resolution of the image data as the published value, and associates and stores the published value and the identification information.
The narrowing-down unit searches the ground resolution storage unit using the identification information included in the candidate data as a keyword, acquires the published value associated with the identification information, and is included in the acquired published value A terrestrial resolution refiner that narrows down the candidate data by comparing the terrestrial resolution with the terrestrial resolution threshold,
A stereo pair image data retrieval system characterized by that.
(Appendix 6)
The stereo pair image data search system according to any one of appendices 1 to 5,
The shooting conditions include a horizontal position accuracy index that defines a horizontal position accuracy index when the image data is captured,
The metadata storage unit includes a horizontal position accuracy index storage unit that stores the horizontal position accuracy index of the image data as the published value in association with the published value and the identification information,
The narrowing-down unit searches the horizontal position accuracy index storage unit using the identification information included in the candidate data as a keyword, acquires the published value associated with the identification information, and acquires the published value A horizontal position accuracy index squeezer that narrows down the candidate data by comparing the horizontal position accuracy index and the horizontal position accuracy index included in
A stereo pair image data retrieval system characterized by that.

(Appendix 7)
A stereo pair image data search method for searching a pair of image data used for creating a three-dimensional terrain model as stereo pair image data,
A procedure for storing the metadata of the geographical conditions related to the geography in which the image data was captured in the metadata storage unit of the database, and storing the published value regarding the shooting conditions in the published value storage unit;
A procedure for specifying the geographical condition and the photographing condition by a user side device;
A procedure for causing the search execution unit to search the metadata storage unit for the metadata that satisfies the geographical condition specified by the user side device, and to output identification information included in the metadata as candidate data;
A procedure for narrowing down the candidate data according to the shooting condition and outputting it as a search result to the narrowing-down section;
A stereo pair image data search method comprising:
(Appendix 8)
The stereo pair image data search method according to appendix 7,
The imaging condition includes a cloud coverage threshold value that defines a cloud coverage rate when the image data is captured,
The metadata storage unit includes a cloud coverage storage unit that stores the cloud coverage of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the cloud coverage storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the acquired metadata A method for searching for stereo pair image data, comprising: performing a cloud coverage narrowing procedure for narrowing down the candidate data by comparing the cloud coverage with the cloud coverage threshold.
(Appendix 9)
The stereo pair image data search method according to appendix 7 or 8,
The shooting conditions include a convergence angle threshold that defines a convergence angle of the stereo pair image data used when creating the three-dimensional terrain model,
The metadata storage unit includes an azimuth / elevation angle storage unit that stores the metadata of the azimuth and elevation when the image data is captured in association with the identification information,
The narrowing-down unit searches the azimuth / elevation angle storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the metadata Calculating the convergence angle using the azimuth angle and the elevation angle, and performing a convergence angle narrowing procedure for narrowing down the candidate data by comparing the calculated convergence angle and the convergence angle threshold,
A stereo pair image data search method characterized by the above.
(Appendix 10)
The stereo pair image data search method according to any one of appendices 7 to 9,
The shooting conditions include an observation date threshold that defines the observation date and time when the image data was shot,
The metadata storage unit includes an observation date storage unit that stores the observation date and time of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the observation date storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the acquired metadata A stereo pair image data search method comprising: performing an observation date narrowing procedure for narrowing down the candidate data by comparing the observation date and time and the observation date threshold.
(Appendix 11)
The stereo pair image data search method according to any one of appendices 7 to 10,
The shooting conditions include a ground resolution threshold value that defines the ground resolution when shooting the image data,
The metadata storage unit includes a ground resolution storage unit that stores the published value and the identification information in association with each other as the published value of the ground resolution of the image data.
The narrowing-down unit searches the ground resolution storage unit using the identification information included in the candidate data as a keyword, acquires the published value associated with the identification information, and is included in the acquired published value Performing a ground resolution narrowing procedure to narrow down the candidate data by comparing the ground resolution and the ground resolution threshold;
A stereo pair image data search method characterized by the above.
(Appendix 12)
The stereo pair image data search method according to any one of appendices 7 to 11,
The shooting conditions include a horizontal position accuracy index that defines a horizontal position accuracy index when the image data is captured,
The metadata storage unit includes a horizontal position accuracy index storage unit that stores the published value and the identification information in association with each other as the published value of the horizontal position accuracy index of the image data,
The narrowing-down unit searches the horizontal position accuracy index storage unit using the identification information included in the candidate data as a keyword, acquires the published value associated with the identification information, and obtains the obtained published value Performing a horizontal position accuracy index narrowing procedure for narrowing down the candidate data by comparing the horizontal position accuracy index and the horizontal position accuracy index included,
A stereo pair image data search method characterized by the above.

2A, 2B Stereo Pair Image Data Retrieval System 10 Retrieval Unit 11 Retrieval Execution Unit 12 Narrowing Unit 12a Convergence Angle Narrower 12b Resolution Narrower 12c Position Accuracy Narrower 12d Cloud Coverage Narrower 12e Observation Date Narrower Device 20 User side device 21 Input unit 21a Geographic condition designation unit 21b Imaging condition designation unit 22 Output unit 30 Database 31 Metadata storage unit 31a Imaging range storage unit 31b Azimuth / elevation angle storage unit 31c Cloud coverage storage unit 31d Observation date and time Storage unit 32 Published value storage unit 32a Ground resolution storage unit 32b Horizontal position accuracy index storage unit

Claims (10)

A stereo pair image data search system for searching a pair of image data used for creating a three-dimensional terrain model as stereo pair image data,
A metadata storage unit that stores metadata of geographic conditions related to the geography in which the image data was captured, and a database that includes a published value storage unit that stores published values related to the shooting conditions;
A user side device including at least an input unit used for designating the geographical condition and the photographing condition;
A search execution unit that searches the metadata storage unit for the metadata that satisfies the geographic condition specified by the user side device, and outputs identification information included in the metadata as candidate data, and the imaging condition A search unit including a narrowing unit that narrows down the candidate data and outputs the result as a search result;
A stereo pair image data retrieval system comprising: The stereo pair image data search system according to claim 1,
The imaging condition includes a cloud coverage threshold value that defines a cloud coverage rate when the image data is captured,
The metadata storage unit includes a cloud coverage storage unit that stores the cloud coverage of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the cloud coverage storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and adds the acquired metadata to the acquired metadata. A cloud coverage refiner is included that narrows down the candidate data by comparing the cloud coverage and the cloud coverage threshold included.
A stereo pair image data retrieval system characterized by that. The stereo pair image data search system according to claim 1 or 2,
The shooting conditions include a convergence angle threshold that defines a convergence angle of the stereo pair image data used when creating the three-dimensional terrain model,
The metadata storage unit includes an azimuth / elevation angle storage unit that stores the metadata of the azimuth and elevation when the image data is captured in association with the identification information,
The narrowing-down unit searches the azimuth / elevation angle storage unit by using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the metadata A convergence angle narrowing device that calculates the convergence angle using the azimuth angle and elevation angle, and narrows down the candidate data by comparing the calculated convergence angle and the convergence angle threshold,
A stereo pair image data retrieval system characterized by that. The stereo pair image data search system according to any one of claims 1 to 3,
The shooting conditions include an observation date threshold that defines the observation date and time when the image data was shot,
The metadata storage unit includes an observation date storage unit that stores the observation date and time of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the observation date storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the acquired metadata Including an observation date and time narrower that narrows down the candidate data by comparing the observation date and time and the observation date and time threshold.
A stereo pair image data retrieval system characterized by that. The stereo pair image data search system according to any one of claims 1 to 4,
The shooting conditions include a ground resolution threshold value that defines the ground resolution when shooting the image data,
The metadata storage unit includes the ground resolution storage unit that stores the ground resolution of the image data as the published value, and associates and stores the published value and the identification information.
The narrowing-down unit searches the ground resolution storage unit using the identification information included in the candidate data as a keyword, acquires the published value associated with the identification information, and is included in the acquired published value A terrestrial resolution refiner that narrows down the candidate data by comparing the terrestrial resolution with the terrestrial resolution threshold,
A stereo pair image data retrieval system characterized by that. The stereo pair image data search system according to any one of claims 1 to 5,
The shooting conditions include a horizontal position accuracy index that defines a horizontal position accuracy index when the image data is captured,
The metadata storage unit includes a horizontal position accuracy index storage unit that stores the horizontal position accuracy index of the image data as the published value in association with the published value and the identification information,
The narrowing-down unit searches the horizontal position accuracy index storage unit using the identification information included in the candidate data as a keyword, acquires the published value associated with the identification information, and acquires the published value A horizontal position accuracy index squeezer that narrows down the candidate data by comparing the horizontal position accuracy index and the horizontal position accuracy index included in
A stereo pair image data retrieval system characterized by that. A stereo pair image data search method for searching a pair of image data used for creating a three-dimensional terrain model as stereo pair image data,
A procedure for storing the metadata of the geographical conditions related to the geography in which the image data was captured in the metadata storage unit of the database, and storing the published value regarding the shooting conditions in the published value storage unit;
A procedure for specifying the geographical condition and the photographing condition by a user side device;
A procedure for causing the search execution unit to search the metadata storage unit for the metadata that satisfies the geographical condition specified by the user side device, and to output identification information included in the metadata as candidate data;
A procedure for narrowing down the candidate data according to the shooting condition and outputting it as a search result to the narrowing-down section;
A stereo pair image data search method comprising: The stereo pair image data search method according to claim 7,
The imaging condition includes a cloud coverage threshold value that defines a cloud coverage rate when the image data is captured,
The metadata storage unit includes a cloud coverage storage unit that stores the cloud coverage of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the cloud coverage storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the acquired metadata A method for searching for stereo pair image data, comprising: performing a cloud coverage narrowing procedure for narrowing down the candidate data by comparing the cloud coverage with the cloud coverage threshold. The stereo pair image data search method according to claim 7 or 8,
The shooting conditions include a convergence angle threshold that defines a convergence angle of the stereo pair image data used when creating the three-dimensional terrain model,
The metadata storage unit includes an azimuth / elevation angle storage unit that stores the metadata of the azimuth and elevation when the image data is captured in association with the identification information,
The narrowing-down unit searches the azimuth / elevation angle storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the metadata Calculating the convergence angle using the azimuth angle and the elevation angle, and performing a convergence angle narrowing procedure for narrowing down the candidate data by comparing the calculated convergence angle and the convergence angle threshold,
A stereo pair image data search method characterized by the above. The stereo pair image data search method according to any one of claims 7 to 9,
The shooting conditions include an observation date threshold that defines the observation date and time when the image data was shot,
The metadata storage unit includes an observation date storage unit that stores the observation date and time of the image data as the metadata, and stores the metadata and the identification information in association with each other.
The narrowing-down unit searches the observation date storage unit using the identification information included in the candidate data as a keyword, acquires the metadata associated with the identification information, and is included in the acquired metadata A stereo pair image data search method comprising: performing an observation date narrowing procedure for narrowing down the candidate data by comparing the observation date and time and the observation date threshold.

Images