KR-102962360-B1 - SERVER AND METHOD FOR ADJUSTING COORDINATE AND SYSTEM INCLUDING THE SAME

KR102962360B1KR 102962360 B1KR102962360 B1KR 102962360B1KR-102962360-B1

Abstract

The present invention relates to a coordinate adjustment server and method capable of generating masking data by performing segment masking on image data, and then performing georeferencing based on the generated masking data, and a system including the same. A coordinate adjustment server may include a data collection module that collects reference data and target data to be adjusted based on the coordinate information of the reference data, a masking module that generates reference masking data and target masking data by performing a segment masking process for each of the reference data and the target data, and a coordinate adjustment module that adjusts the coordinates of the target data based on the reference data, the target data, the reference masking data, and the target masking data.

Inventors

이용주
박현선
황원요
사공용협
이여울
황위주
성정훈
김민정

Assignees

주식회사 메이사

Dates

Publication Date: 20260508
Application Date: 20241224

Claims (20)

A data collection module that collects reference data and target data to be adjusted based on the coordinate information of the reference data; A masking module that generates reference masked data and target masked data by performing a segment masking process for each of the reference data and the target data; and It includes a coordinate adjustment module that adjusts the coordinates of the target data based on the reference data, the target data, the reference masking data, and the target masking data, wherein The above reference masking data and the above target masking data each include data obtained by dividing a plurality of objects included in the above reference data and the above target data, and then masking each of the divided plurality of objects in the form of a predefined shape. The above coordinate adjustment module is, A feature map is extracted for each of the above reference data, the above target data, the above reference masking data, and the above target masking data, and A correlation map is generated by calculating the correlation between feature maps for each of the extracted reference data, target data, reference masking data, and target masking data, and A transformation matrix is generated to perform coordinate transformation of the target data based on the correlation map above, and Generating coordinate adjustment results for the target data based on the transformation matrix generated above Coordinate adjustment server.
In Article 1, The above target data is, including at least one of satellite data, aerial data, and drone data Coordinate adjustment server.
In Article 1, The above masking module is, Generating the reference masking data and the target masking data using a pre-trained masking model Coordinate adjustment server.
In Paragraph 3, The above masking model is pre-trained to output the reference masking data and the target masking data as output data when the reference data and the target data are input as input data. Coordinate adjustment server.
In Paragraph 4, The above masking model is pre-trained to output the reference masking data and the target masking data by dividing the objects included in each of the reference data and the target data, and masking each of the divided objects. Coordinate adjustment server.
In Article 5, The above masking model is pre-trained based on artificial intelligence to output the output data when the input data is input. Coordinate adjustment server.
In Article 6, The above masking model is pre-trained based on a neural network Coordinate adjustment server.
In Article 7, The above masking model is trained based on a Transformer network Coordinate adjustment server.
In Article 1, The above coordinate adjustment module is, Adjusting the coordinates of the target data based on the comparison result between the reference data and the target data and the comparison result between the reference masking data and the target masking data. Coordinate adjustment server.
In Article 9, The above coordinate adjustment module is, A feature extraction unit that generates a reference feature map, a target feature map, a reference masking feature map, and a target masking feature map by extracting the feature map for each of the reference data, the target data, the reference masking data, and the target masking data using a pre-trained feature extraction algorithm, and A correlation calculation unit that generates the correlation map by calculating the correlation between the reference feature map, the target feature map, the reference masking feature map, and the target masking feature map, and A matrix output unit that generates the transformation matrix in the form of matrix data based on the correlation map above, and A control unit comprising: a control unit that generates a coordinate adjustment result for the target data by applying the transformation matrix to the target data to adjust the coordinates of the target data. Coordinate adjustment server.
In Article 10, The above correlation calculation unit is, Generating the correlation map using a matching network that takes the reference feature map, the target feature map, the reference masking feature map, and the target masking feature map as input values. Coordinate adjustment server.
In Article 11, The above matching network is, A weight is assigned to at least one of a plurality of feature points included in the reference masking feature map and the target masking feature map according to a predefined criterion, and Generating the correlation map through matching between a plurality of feature points to which the above weights have been assigned Coordinate adjustment server.
In Article 12, The above weight is determined according to the height of each object corresponding to a plurality of feature points included in the reference masking feature map and the target masking feature map. Coordinate adjustment server.
In Article 10, The above matrix calculation unit is, Generating the transformation matrix from the correlation map using a pre-trained transformation model Coordinate adjustment server.
In Article 14, The above transformation model estimates the transformation matrix from the correlation map through a homography estimation network. Coordinate adjustment server.
In Article 10, The above adjustment unit is, By applying the above transformation matrix to the above target data, preliminary data is generated, which is data in which the above target data is aligned with the above reference data. The coordinate adjustment result is generated by assigning the coordinate information of each pixel of the reference data to each pixel of the preliminary data. Coordinate adjustment server.
Memory for storing at least one instruction; and It includes at least one processor that executes the above at least one instruction, and The above processor is, Collect reference data and target data to be adjusted based on the coordinate information of the reference data, and A segment masking process is performed on each of the above reference data and the above target data to generate reference masked data and target masked data, and Adjust the coordinates of the target data based on the above reference data, the target data, the reference masking data, and the target masking data, wherein The above reference masking data and the above target masking data each include data obtained by dividing a plurality of objects included in the above reference data and the above target data, and then masking each of the divided plurality of objects in the form of a predefined shape. When the processor adjusts the coordinates of the target data, A feature map is extracted for each of the above reference data, the above target data, the above reference masking data, and the above target masking data, and A correlation map is generated by calculating the correlation between feature maps for each of the extracted reference data, target data, reference masking data, and target masking data, and A transformation matrix is generated to perform coordinate transformation of the target data based on the correlation map above, and Generating coordinate adjustment results for the target data based on the transformation matrix generated above Coordinate adjustment server.
In Article 17, The above target data is, including at least one of satellite data, aerial data, and drone data Coordinate adjustment server.
In Article 17, The above processor is, Generating the reference masking data and the target masking data using a pre-trained masking model Coordinate adjustment server.
In Article 19, The above masking model is pre-trained to output the reference masking data and the target masking data as output data when the reference data and the target data are input as input data. Coordinate adjustment server.

Description

Coordinate adjustment server and method and system including the same The present invention relates to a coordinate adjustment server, a method, and a system including the same. Specifically, the present invention relates to a coordinate adjustment server and method capable of performing segment masking on image data to generate masking data, and then performing georeferencing based on the generated masking data, and a system including the same. The content described in this section merely provides background information regarding the present embodiment and does not constitute prior art. In order to utilize data such as satellite imagery, aerial imagery, and drone imagery as spatial information, georeferencing is essential to align the data with coordinate information in an already known coordinate system. Generally, for such georeferencing, a manual method is prevalent in which a human directly selects a Ground Control Point (GCP) and uses it. However, this manual method has problems such as the fact that humans must manually locate and select a point where the reference location can be accurately known, and that there is a possibility of georeferencing failure if GCPs that are not linearly independent are selected. Meanwhile, with the recent advancement of artificial intelligence technology, there is a need to automate such georeferencing using artificial intelligence. FIG. 1 illustrates a coordinate adjustment system according to some embodiments of the present invention. FIG. 2 is a block diagram of a coordinate adjustment server according to some embodiments of the present invention. FIGS. 3 and FIGS. 4 are drawings for illustrating image data according to some embodiments of the present invention. FIG. 5 is a detailed block diagram of a masking module according to some embodiments of the present invention. FIGS. 6 and 7 are conceptual diagrams for explaining the operation of a masking model according to some embodiments of the present invention. FIG. 8 is a diagram illustrating the neural network structure of a masking model according to some embodiments of the present invention. FIGS. 9 and FIGS. 10 are drawings for illustrating the learning and execution steps of a masking model according to some embodiments of the present invention. FIG. 11 is a detailed block diagram of a coordinate adjustment module according to some embodiments of the present invention. FIG. 12 is a diagram illustrating the operation of a feature extraction unit according to some embodiments of the present invention. FIG. 13 is a detailed block diagram of a correlation calculation unit according to some embodiments of the present invention. FIG. 14 is a diagram illustrating the relationship between a correlation calculation unit and a matching network according to some embodiments of the present invention. FIG. 15 is a conceptual diagram illustrating the operation of a matching network according to some embodiments of the present invention. FIG. 16 illustrates the process of a matching network according to some embodiments of the present invention generating a correlation map based on a plurality of feature maps. FIG. 17 is a detailed block diagram of a matrix calculation unit according to some embodiments of the present invention. FIG. 18 is a diagram illustrating the relationship between a matrix calculation unit and a transformation model according to some embodiments of the present invention. FIG. 19 is a drawing for illustrating a conversion model according to some embodiments of the present invention. FIG. 20 is a detailed block diagram of an adjustment unit according to some embodiments of the present invention. FIG. 21 is a drawing for explaining the operation of an adjustment unit according to some embodiments of the present invention. FIG. 22 is a diagram illustrating coordinate adjustment results according to some embodiments of the present invention. FIG. 23 is a block diagram of a coordinate adjustment server according to some other embodiments of the present invention. FIG. 24 is a flowchart of a coordinate adjustment method according to some embodiments of the present invention. FIG. 25 is a detailed flowchart of step (S200) of FIG. 24 according to some embodiments of the present invention. FIG. 26 is a detailed flowchart of step (S300) of FIG. 24 according to some embodiments of the present invention. FIG. 27 is a detailed flowchart of step (S320) of FIG. 26 according to some embodiments of the present invention. FIG. 28 is a detailed flowchart of step (S330) of FIG. 26 according to some embodiments of the present invention. FIG. 29 is a detailed flowchart of step (S340) of FIG. 26 according to some embodiments of the present invention. FIG. 30 is a diagram illustrating the hardware implementation of a coordinate adjustment server according to some embodiments of the present invention. Terms and words used in this specification and claims shall not be interpreted as being limited to their general or dictionary me