KR102070171B1 - System corricting the error position used by image data having the geographic information - Google Patents

Abstract

The present invention relates to an error checking system for location information of digital topography through precise analysis of image information reinforced with geographic information, which completes a highly reliable digital topographic map by modifying a position of a structure image formed in the digital topography to correspond to the actual position, comprising DB and a correction unit.

Description

Digital topographic map location information error checking system through precise analysis of geographic information reinforcement {SYSTEM CORRICTING THE ERROR POSITION USED BY IMAGE DATA HAVING THE GEOGRAPHIC INFORMATION}

The present invention checks the geographic information configured in the image information, verifies whether the target structure image of the digital topographic map is identical to the structure within the image image, and corrects the position of the structural image of the digital topographic map to correspond to the actual position. A digital topographic map location error check system through precise analysis of geospatial-enhanced image information.

In general, image information includes image images collected through aerial photography, and geographic information including coordinate information and attribute information of structures such as roads, rivers, civil structures, buildings, and the like displayed on the image images. However, most of the geographic information was neglected in the process of producing a digital topographic map which constitutes a digital map, and the digital topographic map was produced, modified, and updated based only on the shape of each structure displayed in the image.

On the other hand, the digital topographic map is produced based on video images collected by aerial photography, and field coordinates, which are GPS coordinates measured in the field. However, the image image has a difference in the position and distance of the structure according to the shooting angle, etc. Therefore, the digital topographic map produced based on the image image has a problem that the position and shape of the structure image is changed every time.

In order to solve this problem, a technique of conventionally synthesizing structure images according to field coordinates regardless of image images has been proposed.

However, it was difficult for the operator to manually edit a large number of structure images composed of the image images, and furthermore, there was a limit that the characteristics of the terrain displayed on the image images could not be reflected in the digital topographic map.

In addition, as the actual structure is newly created and destroyed, the numerical topography should be partially updated. In the conventional digital topographic map, the numerical topographic map needs to be manually updated because the structure and the position and the editing have to be changed manually.

Prior Art Documents 1. Patent Registration No. 10-1774878 (announced on September 5, 2017)

Accordingly, the present invention is to solve the above problems, by verifying the geographic information configured in the image information to verify whether the structure target image of the digital topographic map is the same as the structure in the image image, and the actual position of the structure image It is an object of the present invention to solve the problem of providing a digital topographical location information error checking system through precise analysis of image information reinforced with geographic information for correcting the location.

In order to achieve the above object, the present invention,

An image information storage unit for storing and managing image image information; A structure image storage unit for storing the structure shape of the structure of the structure, a region image including the structure image, and storing and managing the structure image for each version; A map information storage unit for storing and managing a field coordinate value of the structure, a reference coordinate value that is a field coordinate of the area image, a numerical coordinate value of the structure image, and general information of the structure; DB, consisting of; numerical data storage for storing data of the digital topographic map;

Analyze the image image to extract the structure and identify nine reference positions located in three rows and three columns in the ground area. The reference coordinate values of neighboring reference positions are the same as the X axis value and the Y axis value. ; A zone information generation module for generating a zone image by drawing the first structure image of the above ground zone-specific structure and inserting the reference position points of the reference positions into the zone image; The reference horizontal axis and the reference vertical axis intersecting the reference position points of one reference position located at the center of the nine reference positions positioned in the three rows and three columns are respectively generated in the zone image, and the reference position points of the reference position located at the center and In addition, a connecting horizontal axis and a connecting vertical axis connecting the reference position points of the eight reference positions positioned around the horizontal and vertical directions, respectively, are generated in the zone image, but the connecting horizontal axis or the non-parallel parallel to the reference horizontal axis or the reference vertical axis line or A reference point analysis module for identifying a reference point on the connection vertical axis and designating it as a flow target point; The connecting horizontal axis and the connecting vertical axis parallel to the reference horizontal axis and the reference vertical axis, respectively, are extended to form a base line in the zone image, and the flow target point deviated from the base line in the zone image to the position of the corresponding intersection point between the base lines. An image pixel flow module for distorting the image in pixel units through a requipy technique to move the image pixel, thereby transforming an image within a reference range of the flow target point in the zone image; By dividing the coordinate lines between the reference position points based on the reference coordinate value of the reference position point, an arbitrary coordinate system is generated in the zone image transformed by the image pixel flow module, and the numerical coordinates of the first structural image based on the arbitrary coordinate system. If the distance between the coordinate lines is different for each section of the reference point, the coordinates are extracted so that all coordinate lines constituting the arbitrary coordinate system are equally spaced, and according to the corrected arbitrary coordinate system, A coordinate data generation module for correcting the position and shape with the second structure image; A CNN module for searching the structure shape of the maximum similarity in the structure image storage unit by analyzing the shape of the second structure image and searching for an image based on a convolutional neural network (CNN); The structure image storage unit retrieves the structure image of the previous version having the numerical coordinate value within the reference range from the numerical coordinate value of the second structure image, and replaces the second structure image with the structure image of the previous version to the third structure image. A structure image correction module for correcting the structure image of the previous version by replacing the second structure image with the structure shape searched by the CNN module if the structure image of the previous version is not detected; The difference between the first angle between the numerical coordinate value of the third structure image and the line segment connecting three or more reference position points and the second angle between the field coordinate value of the structure and the line segment connecting three or more reference position points, respectively A structure position correction module for correcting a numerical coordinate value to a position point on an arbitrary coordinate system corresponding to the second angle when the reference range is exceeded, and moving the third structure image to correct it to a fourth structure image. Topographic Correction

Digital topographical location information error checking system through the precise analysis of the image information reinforced with geographic information, including.

According to the present invention, the geographic information configured in the image information is checked to verify whether the corrected structure image of the digital topographic map is the same as the structure in the image image, and the position of the structural image of the digital topographic map is modified to correspond to the actual position. There is an effect to increase the accuracy of the topographic map.

1 is a block diagram showing an embodiment of an error checking system according to the present invention;
2 is a flowchart showing a process of checking, correcting, and editing a generation error of a digital topographic map based on an error checking system according to the present invention;
FIG. 3 is a flowchart sequentially showing a process after the flowchart of FIG. 2;
4 is a diagram in which an error checking system according to the present invention generates a zone image based on an image image and forms a reference horizontal axis, a reference vertical axis, a connecting horizontal axis and a connecting vertical axis,
5 is a diagram illustrating a state in which an error checking system according to the present invention distorts a zone image.
6 is a view showing an image distortion state of the image pixel flow module configured in the error checking system according to the present invention,
7 is a diagram illustrating a state in which an error checking system according to the present invention generates an arbitrary coordinate system in a structure image.
8 is a diagram illustrating a state in which an error checking system according to the present invention corrects an arbitrary coordinate system and a structure image.
9 is a view showing the shape correction of the structure image in the error checking system according to the present invention,
10 is a view schematically showing a method for correcting a shape of a structure image by an error checking system according to the present invention;
11 is a view showing a state in which the error checking system according to the present invention corrects the position of the structure image by comparing the angle between the reference coordinate value and the numerical coordinate value,
12 is a view schematically showing how the error checking system according to the present invention compares the angle between the reference coordinate value, the numerical coordinate value, and the field coordinate value with each other,
FIG. 13 is a diagram schematically illustrating a position correction inspection of a structure image by an error checking system according to the present invention.

The above-described features and effects of the present invention will be apparent from the following detailed description with reference to the accompanying drawings, whereby those skilled in the art can easily implement the technical idea of the present invention. There will be. As the inventive concept allows for various changes and numerous embodiments, particular embodiments will be illustrated in the drawings and described in detail in the text. However, this is not intended to limit the present invention to a specific disclosure, it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram showing an embodiment of an error checking system according to the present invention, Figure 2 shows a process of checking and correcting and editing the generation of a digital topographic map based on the error checking system according to the present invention 3 is a flowchart showing a process after the flowchart of FIG. 2 in sequence, and FIG. 4 shows that an error checking system according to the present invention generates a zone image based on an image image, and connects a reference horizontal axis, a reference vertical axis, and a connecting horizontal axis. FIG. 5 is a view illustrating a line and a connecting vertical axis line, and FIG. 5 is a diagram illustrating a state in which an error checking system according to the present invention distorts a zone image.

1 to 5, the error checking system 100 of the present embodiment includes an image data storage unit 110, a structure image storage unit 120, a map information storage unit 130, and a numerical data storage unit 140. ) And the digital topographical correction unit 150.

The image data storage unit 110 stores and manages image image information. The image image information includes an image image IM generated by aerial photography, and the image image IM is managed by dividing an area by a position unit. Therefore, a wide range of image images can be formed by synthesizing the image images IM of neighboring regions. The video image information includes a video image IM and location information of the video image IM.

The structure image storage unit 120 includes a structure shape in which the actual shape of the structure S is drawn as it is, a zone image ZI in which one or more structure images DI are arranged, and the structure image DI by version. Manage storage. The structure shape is a sample image simplified by drawing the structure S displayed on the video image as it is. For reference, structures, particularly buildings, etc. may vary only in size, and the overall planar shape may be similar, so that the number of structure shapes is relatively small compared to the number of the same structure (S). Meanwhile, since the zone image ZI is a picture of the image image IM divided into zones, the zone image ZI generates a structure image DI corresponding to the structure S configured in the corresponding image image IM. Configure.

The map information storage unit 130 includes a field coordinate value of the structure S, a reference coordinate value that is a field coordinate of the zone image ZI, a numerical coordinate value of the structure image DI, and all of the structure S. Store and manage information. Since the structure (S) is real, the operator measures the field coordinate value of the corresponding GPS coordinates in the field. In addition, the zone image (Z1) separated by zone unit has more than nine reference positions which are reference points of the corresponding zone, and measures the reference coordinate values which are GPS coordinates for each reference position. Subsequently, the numerical coordinate value is generated based on the arbitrary coordinate system generated in the zone image ZI, and is a position where the structure image DI is located in the arbitrary coordinate system. For reference, the general information for each structure includes position data of pixels linked to itself in the image image.

The numerical data storage unit 140 stores the previous version of the digital topographical information and the new digital topographical information. In the digital topographic map, basic general information as well as general information for each structure image DI are linked to the zone image ZI having the structure image DI. The numerical data storage 140 stores and manages the zone image ZI of both the old version and the new version.

The digital topographical correction unit 150 utilizes the information stored in the image data storage unit 110, the structure image storage unit 120, the map information storage unit 130, and the numerical data storage unit 140, respectively, and generates new information. do. To this end, the digital topographic map correction unit 150 includes an image information analysis module 151, a zone information generation module 152, a coordinate data generation module 153, a reference position analysis module 153 ′, and an image pixel flow module 153. And a CNN module 154, a structure image correction module 155, a structure position correction module 156, and a digital topographic map updating module 157.

A detailed description of each module 151 to 157 configured in the digital topographical correction unit 150 will be described in more detail below.

FIG. 6 is a view showing an image distortion state of an image pixel flow module configured in an error checking system according to the present invention, and FIG. 7 is a view showing a state in which an error checking system generates an arbitrary coordinate system in a structure image. 8 is a view illustrating a state in which an error checking system according to the present invention corrects an arbitrary coordinate system and a structure image, and FIG. 9 is a view illustrating a shape correction state of a structure image in an error checking system according to the present invention. 10 is a diagram schematically illustrating a method of correcting a shape of a structure image by an error checking system according to the present invention.

1 to 10, the error checking system of the present embodiment operates in the following order.

S10; Reference position designation step in video image information

The image information analysis module 151 analyzes the image image IM of the image image information, extracts the structure S, and identifies nine or more reference positions located in the ground area as shown in FIG. . Here, nine reference positions are positioned in at least three rows and three columns in the image image IM. At this time, the reference coordinate values of neighboring reference positions are equal to one of the X axis value and the Y axis value. That is, the 'A' reference position and 'I' reference position neighboring each other in the horizontal direction have the same Y-axis value in the reference coordinate value, and the 'A' reference position and 'Multi' reference position adjacent to each other in the longitudinal direction are the reference coordinate. The X-axis value is the same in the value.

In more detail about the image image analysis, the image information analysis module 151 determines the boundary of the closed zone based on the change of the set color and brightness for each pixel in the image image IM of FIG. The boundary line is drawn along the boundary of the extracted closed section. The boundary line display of the image information analysis module 151 is automatically performed according to a designated process to complete the draft image.

Subsequently, the image information analysis module 151 searches for and extracts the cover page configured in the image image IM to identify the reference position. Since at least nine signs are installed in the ground area, the image information analysis module 151 searches for and extracts at least nine signs from the corresponding image image IM. As described above, the mark is located in at least three rows and three columns, and the reference coordinate value, which is the position coordinate of neighboring marks, should be equal to one of the X axis value and the Y axis value.

For reference, the marker is installed in the ground zone by the operator and is photographed during aerial shooting of the ground zone and displayed on the image image IM.

S20; Geo step verification

The geographic information checking module 151 ′ extracts and checks the geographic information for each structure displayed in the image image IM based on the pixels configured in the previous version of the information, and checks the geographic information on the corresponding pixel of the image image IM. Link.

In more detail, the geographic information verification module 151 ′ first searches the general information of the previous version belonging to the image image IM in the map information storage unit 130. In this case, since the general information stored in the map information storage unit 130 includes location data of pixels linked from the previous version of the image image IM, the geographic information checking module 151 ′ displays the image from the previous version of the general information. The position data of the pixel to be linked to the image IM can be confirmed.

When the general information is searched for and confirmed, the geographic information checking module 151 ′ links the general information to the corresponding pixel of the new image image IM, and processes the geographic information to be formed in the image image IM.

S30; Structure image generation step

The zone information generation module 152 generates the zone image ZI of the corresponding video image IM by allowing the structure image DI of the ground zone-specific structure S to be drawn, and generates a reference position point of the reference position ( SP1 to SP9) are inserted into the zone image ZI. It also processes the collected geographic information to be linked to the pixels of the area image.

In more detail, the zone information generation module 152 provides a general work environment such as a menu that the operator can draw based on the boundary line automatically generated by the image information analysis module 151. The image illustrated in the above environment is generated by the zone information generation module 152 as a zone image ZI as shown in FIG. 4 (b), and the zone information generation module 152 is a reference position point for displaying the reference position. (SP1 to SP9) are inserted into the zone image ZI and edited.

For reference, since the zone image ZI is equally drawn based on the image image IM, the zone information generation module 152 corresponds to the reference position points SP1 to SP9 corresponding to the reference position of the image image IM. Insert and edit it as it is in the zone image (ZI).

Meanwhile, the zone information generation module 152 links the collected geographic information to the pixels of the structure image DI of the zone image ZI so that the zone image ZI constitutes geographic information that is general information of the structure. Therefore, when a structure belonging to the area of the zone image ZI is newly established or destroyed, the geographic information linked to the image image IM and the geographic information linked to the zone image ZI have different forms.

S40; Anchor point analysis stage

The reference position analysis module 153 ′, as shown in (a) of FIG. 5 and (b) of FIG. 5, includes one reference center located at nine reference positions SP1 to SP9 located in three rows and three columns. Check the location. Since the nine reference positions are arranged in three rows and three columns, the eight reference positions are positioned to surround the first second reference position located at the center.

When the reference position located at the center is confirmed, the reference position point analysis module 153 'may include the reference horizontal axis line SX and the reference vertical axis line crossing the second reference position point SP2 which is the position point of the reference position located at the center. SY) are generated in each zone image ZI. Here, the reference horizontal axis SX and the reference vertical axis SY are the references of the arbitrary coordinate system AX (see FIG. 7). This will be described in more detail below.

Subsequently, the reference position point analysis module 153 'moves the reference position points SP1, SP3 to SP9 of the eight reference positions positioned along with the second reference position point SP2 in the lateral and longitudinal directions. The connecting horizontal axis L1 and the connecting vertical axis L2 to L4 are respectively generated in the zone image ZI. In more detail, since the nine reference position points SP1 to SP9 are points formed in the zone image ZI, the reference position analysis module 153 'is a reference position point located in the horizontal or longitudinal direction. Can be connected to each other. For example, the first reference position point SP1 may be connected to the sixth reference position point SP6 and the eighth reference position point SP8 positioned in the lateral direction, and the fourth reference position point may be located in the longitudinal direction. SP4) and the fifth reference position point SP5, respectively. 5 (a) and 5 (b), the connecting horizontal axis line L1 connecting the sixth reference position point SP6 and the eighth reference position point SP8 and the eighth reference position point SP8 are shown. ) And the connecting longitudinal axis L2 connecting the ninth reference position point SP9, and the connecting longitudinal axis L3 and the fourth reference position connecting the first reference position point SP1 and the fourth reference position point SP4. The connecting vertical axis L4 connecting the point SP4 and the fifth reference position point SP5 is shown, respectively.

In addition, the reference position analysis module 153 ′ checks the connection horizontal axis L1 or the connection vertical axis L2 to L4 that is not parallel to the reference horizontal axis SX or the reference vertical axis SY. Since the connection horizontal axis L1 or the connection vertical axis L2 to L4 is a connection line connecting the reference position points SP1 to SP9 with each other, it is determined whether the connection horizontal axis L1 or the connection vertical axis L4 is parallel to the reference horizontal axis SX or the reference vertical axis SY. You can check. For reference, if any connection horizontal axis or connection vertical axis is parallel to both the reference horizontal axis SX and the reference vertical axis line SY, the connection horizontal axis or connection vertical axis line is excluded from the confirmation object.

Subsequently, the reference position point analysis module 153 'is a reference position point on the connection horizontal axis L1 or the connection vertical axis L2 to L4 which is determined to be non-parallel with the reference horizontal axis SX or the reference vertical axis SY. Check (SP4, SP8). Accordingly, the reference position points on the connecting horizontal axis L1 are the sixth reference position points SP6 and the eighth reference position points SP8, and the reference position points on the connection vertical axes L2 to L4 are the first reference position points ( SP1), the fourth reference position point SP4, the fifth reference position point SP5, the eighth reference position point SP8, and the ninth reference position point SP9. However, since the connecting horizontal axis connecting the first reference position point SP1 and the sixth reference position point SP6 is parallel to the reference horizontal axis SX, the first reference position point SP1 located on the connecting horizontal axis line SP1. ) And the sixth reference point SP6 are excluded from the verification target. In addition, since the connecting horizontal axis connecting the fifth reference position point SP5 and the seventh reference position point SP7 is parallel to the reference horizontal axis SX, the fifth reference position point SP5 is also excluded from the confirmation object. In addition, since the connecting vertical axis line connecting the ninth reference position point SP9 and the third reference position point SP3 is parallel to the reference vertical axis line SY, the ninth reference position point SP9 is also excluded from the verification target. As a result, the reference position analysis module 153 'extracts and confirms the fourth reference position point SP4 and the eighth reference position point SP8 and designates it as the flow target point.

S50; Zone Image Distortion Step

The image pixel flow module 153 ″ extends the connection horizontal axis and the connection vertical axis parallel to the reference horizontal axis SX and the reference vertical axis SY, respectively, as shown in FIG. 5B, and the zone image ZI. Base lines AX1, AX1 ', AX2, AX2', AX3, and AX3 'are formed on the substrate.

At this time, the reference horizontal axis line SX and the reference vertical axis line SY via the second reference position point SP2 are also extended to form the basic lines AX1 and AX1 '.

Subsequently, the image pixel flow module 153 "checks the intersection points SP4 'and SP8' between the basic lines AX1, AX1 ', AX2, AX2', AX3 and AX3 ', respectively, and checks the corresponding intersection points SP4' and SP8. ') Is determined as the standard position, so the flow target points (SP4, SP8) separated from the base lines (AX1, AX1', AX2, AX2 ', AX3, AX3') in the zone image (ZI) are the base line (AX1). The image is distorted pixel-by-pixel through the Requipy technique to move to the position of the corresponding intersection point SP4 ', SP8' between AX1 ', AX2, AX2', AX3, and AX3 '.

More specifically, since the flow target points SP4 and SP8 and the intersection points SP4 'and SP8' are different in the arrangement position, the flow target points SP4 and SP8 are sections separated from the standard position. Therefore, in order to move the flow target points SP4 and SP8 to the intersection points SP4 'and SP8' which are standard positions, the image pixel flow module 153 "moves the flow target points SP4 and SP8 in the zone image ZI. Image pixels within the included reference range (Z1, Z2) are moved, where continuity with the image pixels outside the reference range (Z1, Z2) must be maintained, so that the distortion rate is further away from the flow target points (SP4, SP8). As a result, as shown in (b) of FIG. 6, the moving object points SP4 and SP8 move to the right, and the corresponding image is deformed, and the further away from the moving object points SP4 and SP8, the smaller the image deformation of the outline. Lose.

As a result, by moving the flow target points SP4 and SP8 in the zone image ZI, the shape of the structure image DI within the reference ranges Z1 and Z2 from the flow target points SP4 and SP8 is also deformed.

Rikwi pie technology is already widely known technology utilized in the graphic editing software such as Photoshop ^®, a specific algorithm description of the pie rikwi technology will be omitted.

S70; Arbitrary coordinate system generation step

The coordinate data generation module 153 is a coordinate line between the reference position points SP1 to SP9 based on the reference coordinate values x1 and y1 (x2 and y2) (x3 and y3) of the reference position points SP1 to SP9. By dividing (XL, YL1, YL2), an arbitrary coordinate system AX is generated in the zone image ZI deformed by the image pixel flow module 153 ". The interval between the coordinate lines XL, YL1, YL2 is also generated. If the reference position points SP1 to SP9 are different for each section, the arrangement position is corrected so that all coordinate lines XL, YL1, and YL2 constituting the arbitrary coordinate system AX are equally spaced, and the position and shape of the structure image are also shown. Correct in accordance with the corrected arbitrary coordinate system (AX).

In more detail, the reference coordinate values (x1, y1) (x2, y2) (x3, y3) of the reference position points SP1 to SP9 are GPS coordinates and are stored and managed in the map information storage unit 130. The coordinate data generation module 153 retrieves the reference position points SP1 to SP9 of the corresponding image image IM from the map information storage unit 130 to determine the reference coordinate values x1, Check y1) (x2, y2) (x3, y3).

Referring to the drawings in more detail, 'section 1' and 'section 2' are formed between the basic lines AX1, AX1, AX2, AX2, AX3 and AX3 generated as described above, and coordinates The data generation module 153 divides the coordinate lines XL, YL1, and YL2 into respective regions based on the reference coordinate values (x1, y1) (x2, y2) (x3, y3), and then divides the coordinate lines XL, YL1, and YL2 in FIG. As shown in the drawing, an arbitrary coordinate system AX of the zone image ZI is generated. However, since the reference coordinate values (x1, y1) (x2, y2) (x3, y3) of the reference position points (SP1 to SP9) are field coordinates that are GPS coordinates, an error is generated in the zone image Z1 irrelevant to the field coordinates. Is bound to occur. That is, as shown in (a) of FIG. 8, the coordinate line YL1 and the second reference position point SP2 constituted in the section “1” between the first reference position point SP1 and the second reference position point SP2. The distance between the coordinate line YL2 configured in 'section 2' between the third reference position point SP3 may be different in the zone image ZI.

Therefore, the coordinate data generation module 153 may include all coordinate lines XL, YL1, and constituting the arbitrary coordinate system AX when the interval between the coordinate lines XL, YL1, and YL2 is different for each section of the reference position points SP1 to SP9. The position of the coordinate lines XL, YL1, and YL2 is corrected so that YL2 is equidistantly spaced, and as shown in FIG. 8B, the first reference position SP1 and the second reference position SP2 are separated from each other. Section 1 'is corrected to' section 3 'and' section 2 'between the second reference position point SP2 and the third reference position point SP3 is corrected to' section 4 '. In addition, the coordinate data generation module 153 corrects the position and shape of the structure images DI1 and DI2 according to the corrected arbitrary coordinate system AX, and the structure images DI1 'and DI2' as shown in FIG. 8 (b). Correct with).

As a result, the coordinate data generation module 153 may complete the arbitrary arbitrary coordinate system AX in the zone image ZI through editing the reference position.

S70; Steps to Create Numeric Coordinates in Structure Images

When the arbitrary coordinate system AX and the structure images DI1 and DI2 are corrected and generated, the coordinate data generation module 153 checks the midpoint positions of the structure images DI1 and DI2 in the arbitrary coordinate system AX to determine the corresponding structure image ( Digital coordinate values P1 of DI1, DI2) are generated as shown in FIG.

S80; General Information Retrieval Step

The geographic information comparison module 156 'compares the geographic information linked by the zone information generation module 152 to the zone image ZI with the general information of the previous version stored in the map information storage unit 130, and then compares the geographic information with the geographic information. Check whether the corresponding general information is searched.

As described above, since the zone image ZI is a newly generated image based on the new version of the image image IM, the structure image DI and the structure image in the previous version of the zone image ZI are generated. May be different. Therefore, the geographic information of the zone image ZI of the new version may also be different from the geographic information of the zone image of the previous version.

By applying these characteristics, the geographic information comparison module 156 'retrieves the general information linked to the new version of the zone image ZI from the map information storage unit 130 in which the general information corresponding to the previous version of the geographic information is stored. And check whether it is searched or unsearched.

S90; Structure image search step

The CNN module 154 analyzes the shape of the structure image DI generated by the zone information generation module 152 or the coordinate data generation module 153 to perform image search based on a convolutional neural network (CNN). The structure image storage unit 120 searches for the structure shape F1 having the maximum similarity.

As described above, the structure shape F1 is a sample image of various shapes collected by collecting the structure S. Therefore, the structure image storing unit 120 stores and manages a plurality of structure shapes F1 having various shapes, and the CNN module 154 has a structure shape most similar to that of the structure image DI according to a known deep learning image retrieval technique. Search for (F1).

Subsequently, the structure image correction module 155 stores the previous structure image having the numerical coordinate values P1 of the structure image DI extracted by the coordinate data generation module 153 and the numerical coordinate values within the reference range. The structure image (DI) is replaced with the previous structure image as shown in FIG. 9 by searching in the region image of the previous version stored in 120, but if the previous structure image is not found, the CNN module 154 searches for the structure shape (F1). Replace the structure image (DI).

In more detail, the zone image ZI, which is the base image of the digital topographic map, is newly generated and updated for each version according to the change of the structure S and the corresponding terrain, thereby creating a new version of the zone image ZI. If you do, you can refer to the zone image from the previous version. Therefore, before correcting the shape of the structure image DI of the new version of the zone image ZI, the structure image of the zone image of the previous version may be retrieved. The structure image correction module 155 retrieves a structure image having a numerical coordinate value P1 of the structure image DI and a numerical coordinate value located within a reference range in order to retrieve the structure image from the previous region image. Search on.

For reference, the previous structure image searched by the structure image correction module 155 may be a structure S different from the structure image DI of the current zone image ZI. In the structure image, the CNN module 154 compares the structure image DI of the zone image ZI with the shape and color, and checks the matching ratio between the images. As a result of the check, if the matching ratio between the previous structure image and the structure image DI of the zone image ZI corresponds to a reference range, the structure image correction module 155 may determine the structure image of the previous structure image and the zone image ZI. It is assumed that DI) is an image of the same structure S and replaced. However, if the coincidence ratio between the structure image DI of the previous structure image and the zone image ZI is out of the reference range, the structure image correction module 155 may be configured as the structure shape F1 searched by the CNN module 154. Replace image DI.

The CNN module 154 is a well-known algorithm for calculating an image retrieval and a matching rate between two images, and may be applied to the error checking system 100 of the present invention. Since a detailed description of the CNN is already known public technology, a description thereof will be omitted here.

S100; Structural Image Correction Step

When the structure image located in the above condition is found, the structure image correction module 155 considers the structure image as the same as the structure image DI in the new zone image ZI, and views the structure image DI as the previous version. Create a new structure image (DI ') by replacing it with a structure image.

However, if the structure image located in the above condition is not retrieved, the structure image correction module 155 replaces the structure image DI with the structure shape F1 searched by the CNN module 154. For reference, the structure image correction module 155 edits the structure shape F1 according to the arrangement angle and size of the structure image DI as shown in FIG. 10, and if the structure image located in the condition is not retrieved, the edited structure shape Replace the structure image DI with (F1).

In addition, when the geographic information comparison module 156 ′ searches general information corresponding to the geographic information, the structure image correction module 155 searches for and replaces the structure image storage unit of the previous version. That is, if the geographic information comparison module 156 'retrieves geographic information linked to the structure image DI from the general information, the structure image DI may be regarded as the same as the structure image of the previous version.

In addition, when the geographic information comparison module 156 'does not retrieve general information corresponding to the geographic information, the structure image correction module 155 substitutes and corrects the previous structure image with the structure shape searched by the CNN module 154. do. That is, if the geographic information comparison module 156 'does not retrieve geographic information linked to the structure image DI from the general information, the structure image DI may be regarded as an image of the newly constructed structure.

As shown in FIG. 9, since the structure image of the previous embodiment has been retrieved, the structure image correction module 155 ignores the structure shape F1 searched by the CNN module 154 and constructs the structure image as the structure image of the previous version. The image DI was corrected to generate a new structure image DI '.

For reference, the new structure image DI 'is arranged at its midpoint according to the numerical coordinate value P1.

11 is a view showing a state in which the error checking system according to the present invention corrects the position of the structure image by comparing the angle between the reference coordinate value and the numerical coordinate value, Figure 12 is a reference coordinate in the error checking system according to the present invention Figure is a schematic diagram showing a comparison between the angle between the value and the numerical coordinates and the field coordinates.

A description with reference to FIGS. 1 to 12 is as follows.

S110; Structure image position correction step

The structure position correction module 156 may include a first angle a1 between the numerical coordinate value P1 of the new structure image DI 'and the line segments R1 to R3 connecting the three or more reference position points SP1 to SP3, respectively. , a2) and the second angle between the field coordinate value P2 of the structure of the new structure image DI and the line segments R1 'to R3' connecting three or more reference position points SP1 to SP3, respectively. When the difference of b2) exceeds the reference range, the numerical coordinate value is corrected to the position point P2 on the arbitrary coordinate system AX corresponding to the second angles b1 and b2 and the new structure image DI is moved.

In more detail, in the structure S of the structure image DI ', the field coordinate values identified through the GPS coordinate measurement are set, and the reference position points SP1 to SP3 are also the reference coordinate values x1, y1) (x2, y2) (x3, y3). Therefore, between the field coordinates of the structure S and the respective line segments R1 'to R3' connecting the reference coordinate values (x1, y1) (x2, y2) (x3, y3), the second angles (b1, b2) Is calculated. Similarly, the first angles a1 and a2 are also defined between the numerical coordinate values P1 of the structure image DI 'formed in the zone image ZI and the respective line segments R1 to R3 connecting the reference position points SP1 to SP3. ) Is calculated. Therefore, the first angles a1 and a2 in the zone image ZI should be substantially the same as the second angles b1 and b2 in the actual zone.

To this end, the structure position correction module 156 displays the position points P2 and P2 'satisfying the second angles b1 and b2 in the arbitrary coordinate system AX. Two position points P2 and P2 'are identified as shown in FIG. 12 (a), and only one of them is correctly designated by referring to the position of the structure of the second angles b1 and b2. Can be. For example, as shown in FIG. 12A, the P2 location point and the P2 'location point have a relatively large difference in Y value. Therefore, the structure position correction module 156 selects a position point P2 closer to the numerical coordinate value P1 of the corresponding structure image DI 'among the P2 position points and the P2' position points. Specify as a point.

Subsequently, when the difference between the first and second angles a1 and a2 and the second and second angles b1 and b2 exceeds the reference range, the structure position correction module 156 performs the second angle angle (as shown in FIG. 13A). The numerical coordinate value is corrected by the position point P2 on the arbitrary coordinate system AX corresponding to b1, b2). The structure image DI 'is also moved in accordance with the position point P2. However, if the difference between the first and second angles a1 and a2 and the second and second angles b1 and b2 is within the reference range, the structure position correcting module 156 displays the structure image DI ′ in the arbitrary coordinate system AX. It is correctly positioned at, and no further corrections are made.

FIG. 13 is a view schematically illustrating a position correction check of a structure image in the error checking system according to the present invention.

A description with reference to FIGS. 1 to 13 is as follows.

S120 and S130; Calibration check phase and digital topographic map update phase

As shown in FIG. 13B, the digital topographic map updating module 157 includes structure images DI ″, SI1, SI2, and SI3 neighboring each other in the zone image ZI generated by the structure position correction module 156. If the difference between the third angles (c1, c2) of the line segments (R1 "through R3") between the numerical coordinate values of) and the field angle values of the line segments of the corresponding structure (S) is within the reference range, The general information stored in the information storage unit 130 is linked for each structure image to update the digital topographic map and store the numerical information in the numerical data storage unit 140.

In more detail, the digital topographic map updating module 157, like the structure position correction module 156, performs numerical coordinate values of the structure images DI ", SI1, SI2, and SI3 neighboring each other in the zone image ZI. And generate the line segments R1 "to R3" of the liver. The third angles c1 and c2 between the line segments R1 "to R3" thus generated are identified. The structure images DI ", SI1, The field coordinate values of the actual structure S of SI2, SI3) are checked, and the line segment between the field coordinate values is checked. The actual fourth angle between the line segments thus identified is calculated and compared with the third angles c1 and c2 of the numerical coordinate values. As a result of the comparison, if the difference between the actual fourth angle and the third angle (c1, c2) of the numerical coordinate value is within the reference range, the digital topographical update module 157 performs the structure image DI " corrected by the structure position correction module 156. Is determined to be legitimate, and updates and corrections of the digital topographic map are completed by linking the general information of the structure S stored in the map information storage unit 130, namely, the name and address of the structure, for each structure image.

On the other hand, the digital topographic map updating module 157 determines that the arrangement position of the structure image DI ″ is invalid when the difference between the actual fourth square angle and the third square angles c1 and c2 of the numerical coordinate value is out of the reference range. Then, the reference coordinate points SP1 to SP3 based on the arbitrary coordinate system AX are reconfirmed and the above-described generation process is repeated.

As a result, the present invention completes the generation and editing of the structure image DI " in accordance with the reference position of the digital topographical map, and corrects all generation errors of the digital topographical map.

In the detailed description of the present invention described above with reference to the preferred embodiments of the present invention, those skilled in the art or those skilled in the art having ordinary skill in the art will be described in the claims to be described later And it will be understood that various modifications and changes of the present invention can be made without departing from the scope of the art.

100; Error checking system a1, a2; Sacrificial angle AX; Arbitrary coordinate system
AX1, AX1 ', AX2, AX2', AX3, AX3 '; Baseline b1, b2; 2nd angle
c1, c2; Second square angles DI, DI1, and DI2; Structure image
F1; Structure L1; Connecting axis
L2, L3, L4; Connecting longitudinal axis P1; Numeric coordinate value P2; Field coordinate value
R1 to R3; Line segment SP1 to SP9; Location point
SP4 ', SP8'; Crossing point SX; Reference axis SY; Reference vertical axis
XL, YL1, YL2; Coordinate line

Claims (1)

An image information storage unit for storing and managing image image information; A structure image storage unit for storing and managing a structure shape of the structure shape, a region image including a structure image, and storing the structure image for each version; A map for storing and managing general information for each structure including a field coordinate value of a structure, a reference coordinate value that is a field coordinate of the zone image and a reference position, a numerical coordinate value of the structure image, and a pixel position linked to the image image. An information storage unit; DB, consisting of; numerical data storage for storing data of the digital topographic map;
Analyze the image image to extract the shape of the structure and identify nine reference positions located in three rows and three columns in the ground area, wherein the reference coordinate values of neighboring reference positions are the same as the X axis value and the Y axis value. Analysis module; A geographic information verification module for extracting and confirming geographic information for each structure displayed on the video image based on pixels configured in the previous version of the information, and linking the geographic information to the corresponding pixel of the video image; Create a zone image by drawing the first structure image of the ground zone-specific structure, process the collected geographic information to be linked to the pixel of the zone image, and insert the reference point of the reference position into each zone image. Zone information generation module; The reference horizontal axis and the reference vertical axis intersecting the reference position points of one reference position located at the center of the nine reference positions positioned in the three rows and three columns are respectively generated in the zone image, and the reference position points of the reference position located at the center and In addition, a connecting horizontal axis and a connecting vertical axis connecting the reference position points of the eight reference positions located in the circumferential direction and the longitudinal direction, respectively, are generated in the zone image, but the connecting horizontal axis or non-parallel to the reference horizontal axis or the reference vertical axis line or A reference point analysis module for identifying a reference point on the connection vertical axis and designating the reference point as the flow target point; The connecting horizontal axis and the connecting vertical axis parallel to the reference horizontal axis and the reference vertical axis, respectively, are extended to form a base line in the zone image, and the flow target point deviated from the base line in the zone image to the position of a corresponding intersection point between the base lines An image pixel flow module for distorting the image in pixel units through a requipy technique to move the image pixel, thereby transforming an image within a reference range of the flow target point from a zone image; By dividing the coordinate lines between the reference position points based on the reference coordinate value of the reference position point, an arbitrary coordinate system is generated in the zone image transformed by the image pixel flow module, and the numerical coordinates of the first structural image based on the arbitrary coordinate system. If the distance between the coordinate lines is different for each section of the reference point, the coordinates are extracted so that all coordinate lines constituting the arbitrary coordinate system are equally spaced, and according to the corrected arbitrary coordinate system, A coordinate data generation module for correcting the position and shape with the second structure image; A geographic information comparison module for comparing the geographic information linked to the area image by the zone information generation module with previous information stored in a map information storage unit and checking whether or not the general information corresponding to the geographic information is searched; A CNN module for retrieving a structure shape having a maximum similarity in a structure image storage unit by analyzing a shape of the second structure image and searching for an image based on a convolutional neural network (CNN); When the coordinate information of the second structure image has a numerical coordinate value within a reference range or the geographic information comparison module searches for general information corresponding to the geographic information, the second structure image is searched in the structure image storage unit. If a structure image having a numerical coordinate value within the reference range is not retrieved or the geographic information comparison module does not retrieve all the information corresponding to the geographic information, the CNN module searches for the second structural image. A structure image correction module that replaces the structure shape with a third structure image; The difference between the first angle between the numerical coordinate value of the third structure image and the line segment connecting three or more reference position points and the second angle between the field coordinate value of the structure and the line segment connecting the three or more reference position points, respectively A structure position correction module for correcting a numerical coordinate value to a position point on an arbitrary coordinate system corresponding to the second angle when the reference range is exceeded, and for moving the third structure image to correct it to a fourth structure image. Topographic Correction
Digital topographic location information error checking system through a precise analysis of the image information reinforced with geographic information, characterized in that it comprises a.

Images