RU2611602C1 - Method for aerospace images selection from databases based on user criteria and forming of terrain coverage based on these images - Google Patents

Abstract

FIELD: physics, communication.

SUBSTANCE: invention refers to the field of research in the geo-information databases, containing materials of remote sensing, and can be used in data processing, calculation and solution of analytical and photogrammetric tasks. According to the method, the characteristics (metadata) of space images (SI) stored in the database are compared with criterial values with subsequent determination of SI, providing the best performance for the considered area. Structures of mosaic area coverage are plotted on the basis of comparison results, and SI intercrossing. A vector frame is configured for each fragment. The part of SI, included into the mosaic are coverage is cut out. The cut out SI fragments are downloaded to the user workplace.

EFFECT: invention provides improved operational and technical characteristics of aerospace images selection from databases.

3 dwg

Description

The invention relates to the field of functionally oriented search in geoinformation databases containing materials for remote sensing of the Earth, and can be used in data processing, solving computational-analytical and photogrammetric problems. The proposed method allows, on the basis of user criteria, to select the necessary aerospace images (AQI) from the database (s) and, based on them, to form a mosaic coating of the required area that best suits the requirements of the user's solution.

There is a method of constructing and using web directories for focused search (patent RU No. 2382400, IPC G06F 017/30 of 02/20/2010), in which each source of information (site) is assigned a weight that shows the relevance of the site in relation to the thematic section and / or user characteristics.

The disadvantage of this method is the impossibility (or high complexity) of setting the weight coefficient for the possibility of using AKI in solving a specific problem.

There is a method of information retrieval (options) and a computer system for its implementation (patent RU 2506636, IPC G06F 17/30 of 02/10/2014), in which data is searched based on a verbal description of the task by the user.

The disadvantage of this method is the use of headings, as well as the need to convert the verbal description of the task to a formal form.

There is a method of forming the structure of aggregated data and a method of searching for data by means of the structure of aggregated data in a database management system (patent RU 2433467, IPC G06F 17/30 of 11/10/2011), in which the desired information is sorted due to the generated structure of aggregated data and its dynamic update.

The disadvantage of this method is the need for the formation of aggregated data, as well as the difficulty of applying to the search for AKI.

The known method implemented in the utility model of the information retrieval system (patent RU No. 110847, IPC G06F 17/30 of 11/27/2011), which offers a search in a database containing information about the source addresses of information resources.

The disadvantage of this method is the lack of the possibility of forming a mosaic coating of the desired territory, which maximally satisfies the requirements of the user.

There is a method and device for searching metadata using metadata indices (patent RU No. 2304805, IPC G06F 17/30 of 08/20/2007), in which a key is searched for by the index structure of multi-keys containing multi-key values and identifying information for metadata corresponding to the value multi-keys, moreover, multi-keys are structured by means of a combination of predefined metadata fields, after which a metadata fragment is extracted using the found key.

The disadvantage of this method is the need for key generation, which with a small number of fields in the metadata database is not advisable.

The closest in essence to the claimed invention is a method of searching for data about objects and in various databases and a system for its implementation (patent RU 2396593, IPC G06F 17/00 of 08/10/2010), including the formation of a data packet based on the entered parameters custom queries, such as: search attributes and conditions; user identification using an authorization device; creation of a user session context object using the context management device in which the search query parameters are saved and search queries are generated using the user request generation device based on metadata describing the type, composition, structure and ownership of the databases, as well as information, received from the configuration management device specified in the generated data packet of the user request; determination of the set of databases involved in the search, taking into account the level of user access; performing a search for elements common to all databases of classifiers and determining the set of identifiers corresponding to them for each database involved in the search, after which for each of the set of databases a series of queries is performed to search for objects of each type and a list of identifiers of found objects satisfying the search conditions ; obtaining a list of identifiers of found objects that satisfy the conditions of a search query for each database, which is transformed into a single information format and stored in the context object of the user session, while performing a sequence of queries to search for objects of each type, information about which is contained in the database, attribute values objects that are distributed across several tables are collected in one flat denormalized representation of the data taking into account the plurality of attribute values Comrade, containing besides the search attribute values also object identifiers.

The disadvantage of the prototype is that in it all the data corresponding to the user's request is downloaded (copied) to the user's workplace without an assessment of their redundancy. Therefore, when using the prototype for the formation of AKI coverage of the required area, the direct participation of the operator in this process is necessary, namely: sequential extraction from the database of available images of the desired area, a visual assessment of their characteristics and a decision on the possibility of their intended use. The subjectivity, lack of sufficient experience among some operators determines that it is not always possible to quickly obtain the required type of coverage of the AKI terrain with the required characteristics. Sometimes, such an organization of the choice of AKI leads to a significant redundancy of information, namely duplicate coverage of the same area. In addition, even if the database contains 7-10 different images of the required area, the choice of AKI can take up to several tens of minutes. With the number of such AKIs reaching 30 or more, the procedure becomes extremely long and laborious. In addition, copying redundant information to the user's workplace takes extra time.

The aim of the invention is to increase the operational and technical characteristics of the choice of aerospace images from databases, including reducing the time to form the required area with aerospace images and eliminating duplication (redundancy) of coatings.

The claimed benefits are provided by the introduction of new operations: comparing the values of the characteristics (metadata) of the AKI stored in the database with the criterion values and the subsequent determination of the AKI having the best characteristics in the considered area; building the structure of the mosaic coverage of the terrain based on the results of comparison and mutual intersection of the AKI, forming a vector frame for each fragment; cutting out the part of the AKI that is included in the mosaic coverage of the area; uploading to the user's workplace only cut AKI fragments.

A comparative analysis of the technical solution with the method selected as a prototype shows that the inventive method is characterized by new operations, such as: checking the completeness of coverage of a given area with selected AKI; exclusion of excess AKI from the list; formation of vector frames of the informative part of fragments in accordance with the priorities of the characteristics of AKI and their values; cutting fragments from AKI according to the obtained vector frames.

Thus, the claimed technical solution meets the criteria of the invention of "novelty."

An analysis of the known technical solutions in the studied area and in related areas allows us to conclude that the operations introduced are known. However, introducing them into the method of selecting aerospace images from databases according to user criteria and forming a site coverage based on them in this sequence gives this method new properties. The operations introduced are carried out in such a way that they can significantly reduce the formation time of the mosaic coverage of the terrain, eliminate the redundancy of this coverage, which, in turn, allows you to uniquely determine the coordinates of the objects depicted on fragments of AKI included in the mosaic coating.

Thus, the technical solution meets the criterion of "inventive step", as it for the specialist does not explicitly follow from the level of development of technology.

The technical solution can be used in geographic information systems containing a large number of aerospace images with different and close values of the characteristics for the same terrain, to perform photogrammetric processing based on the use of reference images, terrain analysis and other tasks.

Thus, the invention meets the criterion of "industrial applicability".

We will consider the task of selecting aerospace images from databases according to user criteria and forming a coverage of a terrain plot as forming a redundant coverage of a given terrain with images whose characteristics correspond to the requirements of the user's task.

By a non-redundant coating of a given area, we mean a mosaic composed of its multi-temporal aerospace images, providing full coverage of its territory. Each of the images in the mosaic, providing coverage of a certain fragment of a given area, must meet the requirements of the user's task. In the extreme case, such a mosaic may consist of a single image providing full coverage of a given area.

As a necessary condition for the formation of loss-free coverage, we will consider the presence in the database of such AKI, which together provide complete coverage of a given area. As a sufficient condition - the existence of at least one variation of the AKI composition in the mosaic coating of the desired area, in which each image according to its characteristics meets the requirements of the photogrammetric processing.

In FIG. 1 presents a General block diagram of the process of selecting from databases according to the criteria of the user of aerospace images necessary for the formation of a redundant coverage of a terrain,

in FIG. 2 is a flowchart of the formation of a vector frame of informative parts of aerospace images that will be part of the redundant coverage of a terrain,

in FIG. 3 is a block diagram of the formation of a redundant coverage of a site area.

Let there be a database containing a set of D images of the earth’s surface with a power of M. Each image from D corresponds to metadata, the entire set of which contains a set

Each image from D is characterized by a set of G _i , i∈ [1, M] points of terrain covered by AKI data, each of which is assigned geodetic coordinates.

, где U - количество используемых характеристик. При этом будем считается, что изображение, выбираемое в качестве опорного, удовлетворяет требованиям выполняемой задачи пользователя, если значения каждого элемента w_u множества W окажутся в соответствующих интервалах допустимых значений

, задаваемых пользователем.Other characteristics of AKI are presented in the form of many

where U is the number of characteristics used. In this case, it will be considered that the image selected as the reference one satisfies the requirements of the user’s task if the values of each element w _{u of the} set W are in the corresponding intervals of admissible values

defined by the user.

It is required for the formation of a redundant coverage of a given area of terrain with aerospace images that meet the requirements of the user’s task to be performed, find the subset D _{o of the} set D for which the condition

- значение u-й характеристики i-го АКИ.where G ₀ - the set of points of the terrain covered by a mosaic of images composed of D _o , which correspond to the geodetic coordinates; G _T - the set of points within the boundaries of the landfill that limits the required area, each of which corresponds to the geodetic coordinates;

- the value of the u-th characteristic of the i-th AKI.

The method includes the following operations:

1. The user assigns the boundaries of the site of the locality G _T for which it is planned to solve any problems.

2. The user setting specific values for the criteria (2) based on the requirements for solving the problem by the user, as well as the sequence of evaluating the characteristics of the AKI. The highest priority is assigned to the characteristics that have a decisive influence on the solution of the problem by the user.

3. Search in the database (s) of the data and select AKI partially or fully covering the area specified in paragraph 1.

.4. Checking the images selected in step 2 for compliance with user criteria

.

5. The formation of a set of AKI D _n corresponding to the requirements of solving the user problem. Those AKIs for which the values of the characteristics lie in the corresponding allowable intervals are considered as potentially suitable and are included in the set D _n . Other images are excluded from consideration.

6. Checking the completeness of coverage of a given area with images from D _n . Verification of completeness of coverage is carried out by checking the fairness of the expression

where n = 1, ..., N, N is the number of AKI in the set D _n .

In other words, if the set of terrain points within the boundaries of the polygon bounding the desired area is included in the combined set of points within the boundaries of the polygons bounding the terrain covered by the selected images, the transition to the next operation is carried out.

If expression (3) turns out to be incorrect, a return to step 1 is carried out to determine the type of the required partial coverage of a given area and the boundaries of its fragments, for which a search will be made for redundant coatings.

7. Exclusion from D _{n of} excess AKI, resulting in a break-even coating D _o and a mosaic coating of AKI of a given area. For these purposes, an empty set D _{o is formed} . Exclusion of excess AKI is carried out on the basis of enumeration of images from D _n .

8. A polygon G _T is formed corresponding to a part of a given plot of terrain not yet covered by AKI already included in D _o

where Δ is the symmetric difference of G _T and G _n .

.For the first AKI of

.

9. Assessment of the type of formation of the redundant coverage of the terrain specified by the user.

.10. If the priority is the area of the resulting fragments, then from D _n is selected AKI having the maximum coverage area of a given area

.

для решения задачи пользователем. Если наиболее значение приоритетной характеристики у всех АКИ из D_n одинаково, то выбирается АКИ, имеющее наилучшее значение характеристики, обладающей более низким приоритетом.11. If the priority is to obtain AKI fragments with the best values of the characteristics, then from D _{n the} AKI having the best value of the characteristic with the highest priority is selected

to solve the problem by the user. If the highest value of the priority characteristic for all AKIs from D _{n is the} same, then the AKI having the best value of the characteristic with a lower priority is selected.

12. The AKI selected in clauses 10 and 11 is recorded in D _o .

13. A vector frame of the informative part is formed for the selected AKI - that part of the AKI that will be used by the user in solving the current problem.

. Операции 8-14 выполняются до тех пор, пока не выполнен перебор всех АКИ из D_n или требуемый участок местности полностью покрыт АКИ из D_n.14. Evaluation that the required area is completely covered by AKI

. Operations 8-14 are performed until all AKIs from D _n have been enumerated or the required area has been completely covered by AKIs from D _n .

15. An assessment of whether D _{o is formed} . If not, then return to paragraphs 1 and 2 to adjust the search conditions or its termination. If so, go to step 16.

16. Cutting out fragments from the selected AKI (D _o ) according to the vector frame obtained in paragraph 13 and including them in the image mosaic.

17. Unloading AKI fragments and vector frames to the user's workstation.

Thus, when using the proposed method, an increase in the operational and technical characteristics of the choice of aerospace images from databases is achieved, including reducing the time for forming the required area of the terrain with aerospace images and eliminating duplication (redundancy) of coatings. This excludes a number of user operations for the user to view each AKI matching his criteria and select the best ones. Elimination of redundancy and cutting out AKI fragments with an informative part can reduce the time it takes to upload data to the user's workstation.

Claims (1)

A method for selecting aerospace images (AKI) from databases according to user criteria and forming a coverage of an area based on them, characterized in that for a given area of terrain a mosaic coating is built on the basis of determining by user criteria that part of the AKI that will be used to solve the problem by the user for this proposed method of: forming a search query to the metadata database comprising identifiers and values AQI characteristics, with the desired site boundaries specifying areas G _T, The values and sequence of the search criteria (priority) estimation AQI characteristics; carry out a search in the database (s) of AKI data partially or fully covering a given area; selecting from the found AKI matching user criteria; assess the completeness of coverage of a given area; exclude excess AKI; if the coverage is not complete, then the search criteria are adjusted, otherwise a mosaic coating is formed by cutting out fragments from the selected images of a given area; for each fragment, a vector frame of the informative part of the image used to solve the problem by the user is formed, so that the combination of the vector frames of the informative part of all fragments is G _T without duplicate coatings by different fragments of the same area; unload AKI fragments and vector frames of informative parts to the user's workplace.

Images