KR101475895B1 - Image processing system recognizing point and auto-correcting for image update - Google Patents

Abstract

The present invention relates to a processing system which automatically checks a difference between an actual site and the position of various object images displayed in an image and corrects the checked difference and, more specifically, to an image processing system recognizing and automatically correcting a point for the update of an image, in order to quickly process a corresponding ground object image with a high degree of accuracy by automatically recognizing an arrangement point of a ground object image which needs a correction and automatically identifying the degree of the correction. An image having high accuracy is completed by: automatically recognizing a corresponding ground object image and the arrangement point of the object image in an image, correcting the recognized ground object image and the recognized arrangement point, extracting the image of a shading section from a different image, and synthesizing and supplementing the extracted image in order to expose the shading section, which is shaded by the ground object image of an inclined shape, to be visible; and correcting the ground object image into a right position in the image by comparing a position, in which the ground object image is positioned in the image, with actual GPS coordinates of a ground object which is an object of the ground object image, if a primary correction of the ground object image is completed.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a video image processing system,

The present invention relates to a processing system for automatically checking whether a position of various images posted on a video image is different from an actual site and correcting the same, and more particularly, The present invention also relates to a video image processing system of a point recognition and automatic correction method for updating a video image which is recognized automatically as a video image and automatically recognizes the degree of correction and quickly processes the ground image with high accuracy.

The background of the map is typically done through aerial photographed images. In other words, the airplane located at a predetermined altitude captures the ground, and secures a video image as shown in Fig. 1 (an image showing a state of a conventional aerial photographing map). Of course, since the shooting area is limited according to the altitude, it is necessary to acquire a plurality of image images for the same section for producing a map for a wide range, and to perform separate editing and image processing tasks do.

However, since the video image is photographed from an aircraft located at a certain altitude from the ground, except for the ground water located in the vertical direction of the aircraft, other adjacent grounds must be photographed including the side surface. For reference, the "Gangnam Cheil Building" under the white underline located at the upper left of FIG. 1, the "Pull Rim Industry" building under the white underline at the upper right, and the "Meritz Tower" under the white underline located at the lower left, (The plane portion), but not the side.

Meanwhile, in order to produce a map, a plurality of image images must be connected to each other as described above. In this process, the image processing task of partially applying the image images photographed at other locations is performed. Finally, as can be seen from the three buildings presented above, the maps produced by the conventional mapping method are displayed on the same direction but the three neighboring buildings are tilted in a completely different direction.

Therefore, the user has difficulty in interpreting the map of an unfamiliar area, and thus, it becomes inconvenient to use the map.

In order to solve such a problem, a conventional technology (Patent Registration No. 10-0967448, published on July 1, 2010) for editing and processing a ground-related image such as various buildings has been proposed. The image processing technique disclosed in the prior art is to edit the ground image generated in the 3D format by the photographing angle so as to be expressed in 2D format by modifying the position of the plane portion.

However, since the shading section correction performed in the image processing process of the related art is to edit and process another image image in which the actual image of the corresponding shade portion is captured, the operator manually searches for the corresponding image image, The process of editing and matching the size and resolution was required. That is, the video image search and editing for the shade section are manually performed, and thus, the image processing time for a number of ground image images is required.

As a result, there is a problem in that a user can not be satisfied with a video image, because a video image having a ground image is omitted from the conventional technique or the entire ground image can not be processed even if the conventional technique is applied.

Prior Art Document 1. Registration Patent Publication No. 10-0967448 (issued on July 1, 2010)

SUMMARY OF THE INVENTION Accordingly, the present invention has been made to solve the above-mentioned problems, and it is an object of the present invention to provide an image processing apparatus and method, which can adjust a three-dimensional ground water image posted in a slanting shape in a video image, thereby eliminating a shadow region caused by the ground water image, The present invention provides a video image processing system of a point recognition and automatic correction method for updating a video image in which a position of a video image and an actual field are increased.

에 대입해 촬영각(θ)을 연산하고, 제1지상물이미지(20a)의 평면부 폭(w2)과 촬영각(θ)을

에 대입해 수정된 평면부 폭(L2)을 연산하고, 제1지상물이미지(20a)의 수정 전 평면부 이미지를 분리해서 수정된 평면부 폭(L2)에 맞춰 그 크기를 조정하고, 제1지상물이미지(20a)의 측면부 이미지는 제거해서 제2지상물이미지(20a')를 완성하는 영상처리모듈(7);을 포함하는 영상처리시스템에 있어서,
상기 제1지상물이미지(20a)를 포함한 영상이미지 데이터의 GPS좌표를 기초로, 제1지상물이미지(20a)의 지상물을 다른 각도에서 촬영한 제3지상물이미지(20a")를 포함하는 영상이미지 데이터를 이미지DB(1)에서 검색하는 영상이미지 검색유닛(61);
이미지탐색모듈(5)에서 확인한 지상물이미지들의 경계를 기준으로 제1지상물이미지(20a)와 제3지상물이미지(20a")의 형태인 윤곽을 각각 확인하되, 제1지상물이미지(20a)와 제3지상물이미지(20a") 각각의 윤곽에서 평면부와 측면부를 구분하고, 평면부만을 각각 추출해서 제1지상물이미지(20a)의 윤곽으로 하는 제1이미지와, 제3지상물이미지(20a")의 윤곽으로 하는 제2이미지로 완성하는 이미지윤곽 확인유닛(62);
제1지상물이미지(20a)의 평면부와 측면부에 대한 윤곽을 제1윤곽형상으로 하고, 상기 제1윤곽형상의 둘레에 인접한 다른 지상물이미지들의 윤곽을 제2윤곽형상을 확인해서 상기 제1,2윤곽형상의 배치상태를 확인하고, 제3지상물이미지(20a")의 평면부와 측면부에 대한 윤곽을 상기 제1윤곽형상에 대응하는 제3윤곽형상과, 상기 제2윤곽형상에 대응하는 제4윤곽형상을 각각 확인해서, 상기 제2윤곽형상들로 이어진 경계로 형성된 제1구역과, 상기 제4윤곽형상들로 이어진 경계로 형성된 제2구역을 각각 확정하고, 상기 제1윤곽형상을 갖춘 제1구역의 중심과 제3윤곽형상을 갖춘 제2구역의 중심을 맞춰 중첩하며, 상기 제1윤곽형상과 제3윤곽형상의 위치를 비교해서 상기 제1윤곽형상의 평면부와 제3윤곽형상의 평면부 간 이격거리(BD)를 확인하는 한편, 확인한 이격거리(BD) 중 가장 긴 이격거리(BD)를 갖는 제2이미지와 해당하는 제3윤곽형상을 확인하는 지상물이미지 확인유닛(63);
음영부분(D)에 대응하는 실제이미지의 범위를 상기 제1윤곽형상과 제3윤곽형상이 서로 겹치지 않는 범위로 해서 확정하고, 제3지상물이미지(20a")가 포함된 영상이미지에서 상기 범위만큼 실제이미지를 추출하며, 추출한 상기 실제이미지의 해상도, 크기 및 색상을 제1지상물이미지(20a)가 포함된 영상이미지의 해상도, 크기 및 색상에 맞춰서 편집하는 한편, 편집한 상기 실제이미지를 제2지상물이미지(20a')가 포함된 영상이미지의 음영부분(D)에 합성하는 음영부분 처리유닛(64);
영상이미지에 링크된 GPS좌표계를 기준으로 제2지상물이미지(20a')가 위치한 지점과, 제2지상물이미지(20a')의 대상이 되는 지상물의 GPS좌표를 비교해서 차이가 확인되면, 상기 GPS좌표에 해당하는 상기 GPS좌표계의 일지점으로 제2지상물이미지(20a')의 위치를 수정하고, 제2지상물이미지(20a')의 색상 및 명암을 주변이미지의 색상 및 명암과 일치하도록 조정하는 위치수정유닛(65);According to an aspect of the present invention,
An image DB 1 for storing image image data to which GPS coordinates are applied, image height h2 of the image image, and height information h1 of the ground image; An image retrieval module 2 for retrieving data and information of the image DB 1; An output module 3 for outputting image image data retrieved by the image retrieval module 2 through a touch screen 3a having an input function; (P1) and a pair of end points (P2) in a video image outputted to the touch screen (3a), and sets a reference straight line connecting the coordinate values to obtain an initial point P1 ) And the end point (P2) to the range of the road image (30); A predetermined color of a pixel corresponding to the road image 30 is confirmed and a boundary of the pixel range designated by the same color as the designated color of the road image 30 on the reference straight line is inconsistent, A1 to a6 of the first ground image 20a and the first ground image 20a forms a search straight line in a direction covering the road image 30 and outputs the same through the output module 3 It is possible to confirm the designated color of the pixel corresponding to the selected point on the search straight line and confirm the range of the same pixel as the boundary to the boundary of the reference ground image 10a, The reference point b1 to b4 of the reference ground image 10a and the width w2 of the first ground image 20a of the first ground image 20a through the reference points a1 to a6 of the first ground image 20a, An image search module 5 for calculating an image of an object; The center distance d between the reference ground 10 and the ground 20 and the height h2 of the ground and the photographing height h2 of the image are checked

(W2) and the photographing angle ([theta]) of the first ground-based image 20a are calculated

(L2) of the first ground-water image (20a), and adjusts the size of the first sub-surface image (20a) in accordance with the modified plane width (L2) An image processing module (7) for removing a side image of the ground image (20a) to complete a second ground image (20a '),
And a third ground image 20a " obtained by photographing the ground water of the first ground water image 20a from another angle, based on the GPS coordinates of the image image data including the first ground water image 20a A video image retrieval unit (61) for retrieving the video image data from the image DB (1);
The outline in the form of the first ground image 20a and the third ground image 20a " is checked based on the boundary of the ground image images confirmed by the image search module 5, and the first ground image 20a ) And the third terrestrial image 20a "in the outline of the first terrestrial image 20a ", respectively, and extracting only the planar portion and forming the outline of the first terrestrial image 20a, An image contour confirmation unit 62 for completing a second image having an outline of the image 20a ";
The first contour of the first ground image 20a is a first contour, and the contour of other ground images adjacent to the first contour is identified as a second contour, , And confirms the arrangement state of the two contour shapes. The third contour shape corresponding to the first contour shape and the third contour shape corresponding to the second contour shape correspond to the flat portion and the side portion of the third terrestrial object image 20a " And a second region formed by a boundary extending to the fourth outline shapes, respectively, and the first contour shape is determined by the first contour shape, And a third area having a third contour shape, and comparing the positions of the first contour shape and the third contour shape with each other to compare the positions of the first contour shape and the third contour shape with each other, (BD) between the flat part of the contour shape, Interval distance (BD) a second third ground water image check unit 63 to determine the outline form corresponding to the image that has the longest distance (BD) of the;
The range of the actual image corresponding to the shaded portion D is determined to be a range in which the first contour shape and the third contour shape do not overlap with each other, Size and color of the extracted actual image in accordance with the resolution, size, and color of the video image including the first ground image 20a, and outputs the edited actual image as an image A shaded portion processing unit 64 for synthesizing the shaded portion D to a shaded portion D of the video image including the two-ground-water image 20a ';
When the difference between the point where the second ground image 20a 'is located based on the GPS coordinate system linked to the video image and the GPS coordinates of the ground object which is the object of the second ground image 20a' The position of the second ground image 20a 'is corrected to one point of the GPS coordinate system corresponding to the GPS coordinates, and the color and the contrast of the second ground image 20a' A position correcting unit 65 for adjusting the position;

An image editing module 6,

delete

delete

delete

delete

delete

delete

And an image processing system of a point recognition and automatic correction method for updating a video image.

In the present invention as described above, the tilted ground water image posted on the video image is corrected so that the ground water image is corrected, the tilted shadow section is exposed, and the corresponding point of the corrected ground water image is automatically recognized It is possible to complete the video image with high reliability while minimizing the labor of the operator.

1 is an image showing a conventional aerial photograph map,
2 is a block diagram illustrating an image processing system according to the present invention,
3 is a view showing a video image to which a video image processing system according to the present invention is applied,
4 is a flowchart sequentially showing an image processing method according to the present invention,
5 is a view schematically showing an aerial photographing for applying an image processing method according to the present invention,
FIG. 6 is a view showing a state of a ground object corrected according to an image processing method according to the present invention,
7 is an image showing a corrected vertical image according to an image processing method according to the present invention,
FIG. 8 is a block diagram illustrating a configuration unit of the image editing module according to the present invention,
9 is a flowchart sequentially showing the operation sequence of the image editing module according to the present invention,
10 is an image showing a video image processed by the image editing module,
11 is a view showing an outline of a video image checked by the image editing module,
12 is a diagram illustrating a process of extracting a shadow portion by the image editing module,
FIG. 13 is a view schematically showing a state in which contour shapes of a ground water image are compared,
14 is an image showing a video image processed by the image editing module,
15 is a view showing a state before the arrangement point of the ground water image is corrected,
16 is a view showing a state after the modification of the arrangement point of the ground-water image.

BRIEF DESCRIPTION OF THE DRAWINGS The above and other features and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings, It will be possible. The present invention is capable of various modifications and various forms, and specific embodiments are illustrated in the drawings and described in detail in the text. It should be understood, however, that the invention is not intended to be limited to the particular forms disclosed, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.

In addition, the present invention utilizes a part of the system configuration of the above-mentioned prior Japanese Patent No. 10-0967448. Therefore, some of the features of the device configuration described below are those described in Patent No. 10-0967448.

Therefore, the device structure, characteristics and operation relationship described below will be referred to as the contents of the above-mentioned Japanese Patent Application No. 10-0967448, and the structure related to the main features of the present invention will be described in detail in a part of the detailed description .

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings.

FIG. 2 is a block diagram illustrating a video image processing system according to the present invention. FIG. 3 is a diagram illustrating a video image to which the video image processing system according to the present invention is applied. Referring to FIG.

The image processing system according to the present invention includes an image DB 1 for storing image image data of various regions, an image search module 2 for searching specific image image data in the image DB 1, A point selection module 4 for setting a search criterion of a ground image to be corrected by selecting a specific point in the image image outputted by the output module 3, An image search module 5 for searching the first ground image 20a (see FIG. 6 (a)) before the modification, and a second ground image 20a ' An image editing module 6 for applying and updating the existing image data to the image processing module 7 for modifying the first ground image 20a to be corrected .

Since the image DB 1 for storing data and the image retrieval module 2 for retrieving specific data in the image DB 1 are general technologies in which the structure and structure are well known and commonly used, a description thereof will be omitted here .

The output module 3 outputs data retrieved by the image retrieval module 2 to a known and common touch screen 3a having a screen input function. In the present invention, the data is an image image, and the output module 3 is a known and common module for outputting a video image through the touch screen 3a. Of course, the output module 3 has a function of confirming a point where the user touches the touch screen 3a and inputting the information. For reference, the output module 3 can check image data and output it as an image, such as 'Photoshop (a Leicester graphic editor developed by Adobe Systems Incorporated)' and 'Paint' which is a graphic editor of Windows which is a representative operating system of Microsoft Lt; / RTI >

The point selection module 4 sets a reference for searching the first terrestrial image 20a in the video image by the video image processing system according to the present invention, (30).

More specifically, in the vicinity of the reference ground 10 (see FIG. 5) and the modified ground 20 (see FIG. 5), a road with a uniform color of aggregate such as asphalt is located. Since this road is very close to the reference ground water 10 or the modified ground water 20, the first ground water image 20a, in which the side portion of the modified ground water 20 is photographed, It is necessary to occupy the edge portion of the road image 30.

The image processing system according to the present invention applies this characteristic. The point selection module 4 includes a pair of initial point P1 and a pair of end points P2 ) As the road image (30). That is, the point selection module 4 confirms each coordinate value of the initial point P1 and the end point P2 selected by the user and connects them to the reference straight line. The reference straight line thus formed becomes the road boundary, 30) is determined.

The image search module 5 checks the color of the road image 30 whose range and its boundary are confirmed by the point selection module 4 and then checks the designated color of the pixel corresponding to the edge of the road image 30 , The color of the road image (30) is compared with the color of the road image (30). In more detail, the image search module 5 checks the designated color of the pixel corresponding to the reference straight line set in advance, compares the color thus verified with the color of the road image 30, If a period T of a pixel whose color is the same as the color of the image 30 and is determined to be the same as each other is determined, the section T is regarded as the first ground image 20a and is set as a correction target .

Of course, even if the specified color of the pixel corresponding to the first ground-based image 20a and the specified color of the pixel corresponding to the road image 30 coincide with each other, If the same color similar to the color of the road image 30 is consecutively identified to the pixel at a position apart from the reference straight line by more than a predetermined distance from the reference straight line, it is regarded as the first ground image 20a, Setting.

Subsequently, the image processing module 7 modifies the set first ground image 20a, and the image editing module 6 applies the modified second ground image 20a 'to the image image to update , Which will be described in more detail below.

FIG. 4 is a flow chart showing an image processing method according to the present invention, and FIG. 5 is a schematic view of an aerial photographing for applying an image processing method according to the present invention, And is a view showing a state of the ground water corrected according to the processing method.

The image processing method according to the present invention corrects a side portion of a ground object photographed at the time of aerial photographing so that the completed image image can display only the plane of the ground object accurately so that the image image can effectively perform the function of the map .

The image processing method will be described sequentially with reference to the flowchart of Fig.

S11; Selecting Modified Steps

The output module 3 outputs a video image including various ground image images 10a and 20a and the point selection module 4 and the image search module 5 are connected to the first ground image 20a, .

S12; Step of setting the reference point to be modified

When the first ground image 20a is determined, the range occupied by the first ground image 20a in the video image is checked, and the edge of the first ground image 20a is defined as the reference point a1 to a6. . Of course, the edges selected as the reference points a1 to a6 include the edges a3 and a4 located at the boundary portion between the plane portion and the side portion of the first ground image 20a as shown in Fig. 6 (a). To this end, the image search module 5 confirms the designated color of the pixel corresponding to the first ground-water image 20a, thereby determining the boundary of the first ground-water image 20a, And sets the reference points a1 to a6.

The image search module 5 sets the reference points a1 to a6 of the first groundwater image 20a so that the total width w1 of the first groundwater image 20a based on the reference points a1 to a6, And the width w2 of the flat portion.

For reference, the reference points a1 to a6 are set and the image search module 5 confirms the coordinate values for the corresponding points via the pixels, and the first ground image image The total width w1 and the width w2 of the flat portion 20a can be calculated. Here, the total width w1 and the planar portion width w2 will be the total length of the first ground image 20a in the inclined direction and the length of the plane portion.

S13; Select the baseline target

When photographing the ground on an aircraft in flight, the plane of the specific ground object can be accurately photographed as shown in Fig. Of course, since only the plane of the reference ground object 10 can be photographed at 100% and output to the image image, if the user can identify the surrounding road image 30 while being perceived as a plane when checking with the naked eye, Lt; RTI ID = 0.0 > 10a. ≪ / RTI >

On the other hand, it is preferable that the user selects the reference ground object 10 satisfying the above-mentioned condition as a reference object, considering the position of the modified ground object 20 selected as the object of correction. In other words, the reference ground water 10 selects the ground water located on the same straight line as the tilted direction so that the side of the modified ground water 20 is seen, and uses the ground water as a reference target.

Accordingly, the image search module 5 forms a search straight line in a direction in which the first ground image 20a covers the road image 30, and the output module 3 outputs the search straight line to the touch screen 3a So that the user can touch the point to be selected as the reference ground image 10a among the images positioned on the search straight line so as to input the same.

S14; Reference target point setting step

When the reference ground image 10a is selected, the image search module 5 confirms the designated color of the point pixel touched by the user, and determines the range of similar pixels, which are the same as the color, to the boundary of the reference ground image 10a And the corner points are confirmed at the thus determined boundaries to set the reference points b1 to b4.

When the reference points b1 to b4 are set, the image search module 5 calculates the plane width L1 of the reference ground water image 10a based on the reference points b1 to b4 as described above.

S15; Each shooting step

When the reference groundwater image 10a is selected, the image processing module 7, as shown in Fig. 5, picks up the reference groundwater 10 while the airplane being photographed is positioned in the upper right room of the reference ground water 10 .

On the other hand, the image processing module 7 confirms the actual center distance d between the reference ground 10 and the ground 20. The center distance d is calculated by calculating the distance between the center point in the reference points b1 to b4 of the reference ground image 10a and the center points in the reference points a3 to a6 of the plane portion of the first ground image 20a, The value can be obtained by converting the scale of the video image. The image search module 2 searches and confirms the altitude h2 of the aircraft and the actual height h1 of the modified ground 20 at the time of shooting the image in the image DB 1. [

Subsequently, the image processing module 7 calculates the photographing angle [theta] of the corrected ground object 20 photographed on the aircraft. Here, the photographing angle? Refers to the angle of the photographing direction of the camera and the arrangement direction of the modified ground 20. Therefore, the photographing angle [theta] at which the camera of the aircraft photographs the reference ground water 10 while the airplane is located in the room directly above the reference ground water 10 will be '0 degree'.

The following formula (1) is used for the imaging angle (?) Calculation.

S16; Modification target range Operation step

5 and 6, if the camera of the aircraft is not located in the upper right room of the modified ground water 20, the photographed first ground water image 20a is transmitted to the plane portion of the modified ground water 20 The side part is included and output. That is, the area in the image image that the first ground image 20a seems to occupy differs from the area obtained by calculating the planar area occupied by the ground by the corrected ground 20 as the scale of the image.

The first ground water image 20a is obtained by photographing the modified ground water 20 at an angle to the first ground water image 20a during aerial photographing, So that the ground part, which is not actually occupied by the modified ground water 20, is covered by the modified ground water 20. The portion of the image captured in this way is confirmed as a part of the modified ground 20 so that the first ground image 20a is larger than the actual appearance of the modified ground 20 . Of course, such an error may cause the road image 30 adjacent to the first ground image 20a to be hidden from view, so that a user who uses the map based on the image image may feel confused in using the map.

Therefore, in order to solve such a problem, it is necessary to correct the size of the first ground image 20a outputted to the video image, and to correct the video image which has the same effect as when the aircraft is photographed in the room immediately above the ground 20 Go ahead.

To this end, the image processing module 7 substitutes the plane width w2 and the photographing angle? Of the first ground image 20a in the image to be corrected into the equation (2) (L2) when the plane of the screen 20 is photographed in the room immediately above it.

S17; Correction target image correction step

The image processing module 7 completes the second ground image 20a 'corrected in accordance with the corrected plane width L2. The corrected second ground image 20a 'is removed from the side portion of the existing first ground image 20a, and the size of the plane portion is corrected in accordance with the modified plane portion width L2.

Here, the image processing of the second ground image 20a 'by the image processing module 7 proceeds as follows.

First, only the plane portion of the existing first ground image 20a is cut out to obtain an independent plane image. The thus obtained planar image is deformed according to the ratio of the modified wing width L2 of the modified second ground image 20a 'to the width w2 of the planar portion of the existing first ground image 20a, Such a modification can employ a public image modification technique.

For reference, the image transformation technology refers to a technique called 'Paste' (image editing) using 'Photoshop (a Leicester graphic editor developed by Adobe Systems Inc.)', 'Paint', a graphic editor of Windows, The image processing module 7 can adjust the size of an image at a corresponding ratio by determining the ratio of the width and height of the image through the function.

On the other hand, the reference when synthesizing the corrected plane portion by the image editing module 6 is assumed to be the edges a1 and a2 directly facing the reference ground image 10a. This is because the edges a1 and a2 are always fixed positions on the ground as compared with the reference ground surface water image 10a irrespective of the tilted state of the first ground water image 20a.

S18; Image synthesis step

As shown in Fig. 6, when the correction of the second ground image 20a 'is completed, the processing of the shade portion D, in which the existing first ground image 20a is not occupied and output, is required. To this end, the image editing module 6 searches the image DB 1 for another image image in which the actual image of the shaded portion D is photographed through the image search module 2, The water image 20a 'is made to coincide with the size and resolution of the image including the image. For reference, since the video image stored in the image DB 1 is data that performs its function as a digital map, the video image is digital map data to which GPS coordinates are applied. Accordingly, the image editing module 6 confirms the GPS coordinates of the shaded portion D and searches the image DB 1 based on the GPS coordinates to search for another image image in which the entire shaded portion D is normally output have.

If the size and resolution of the actual image match with the resolution of the image, the image editing module 6 converts the actual image of the shaded portion D from the other image to the image of the image including the second ground image 20a ' And synthesized with the shaded portion D to remove the shaded portion D as shown in Fig. 7 (an image showing a vertical image corrected according to the image processing method according to the present invention), and a complete vertical image is output .

The technique according to the present invention for shaded portion (D) processing is described in more detail below.

S19; The vertical image data updating step

When the correction of the second ground image 20a 'is completed, the image editing module 6 inputs the image image data in which the actual image is synthesized to the image DB 1 and updates it.

FIG. 8 is a block diagram illustrating a configuration unit of an image editing module according to the present invention. Referring to FIG.

As described above, in order to recover the shaded portion D generated by the conventional image editing module 6 in the process of image processing of the ground image, a manual operation of the operator is required in some processes. However, the image editing module 6 'according to the present invention may be configured such that the third ground image 20a' ', which is the ground image of the ground water that is the same as the first ground image 20a (see FIG. 10) ) Is automatically searched to check the actual image of the shaded portion (D), and the actual image thus confirmed is automatically synthesized with the corresponding image image under image processing, so that the second image processing So that editing of the ground image 20a '(see FIG. 14) and the shaded portion D is automatically and quickly performed.

To this end, the image editing module 6 'according to the present invention includes a video image search unit 61 for searching a video image in the image DB 1 in cooperation with the image search module 2, An image outline confirmation unit 62 for confirming the outline of the images, and a third ground image image 20a "corresponding to the first ground image 20a in the outline-determined video image of the ground water images (See Fig. 12) between the plane portion of the first ground-water image 20a and the plane portion of the third ground-based image 20a " A shaded portion image processing unit 64 for identifying the actual image corresponding to the shaded portion D and compositing the processed image to the processed ground image and completing the second ground image 20a ' And the position of the second ground image 20a 'on the basis of the coordinate information linked to the video image It includes a prescribed position modifying unit (65).

The image editing unit 63 and the image correction unit 65 constitute the image image search unit 61 and the image contour confirmation unit 62 and the ground image image confirmation unit 63 and the shade part processing unit 64 and the position correction unit 65 constituted by the image editing module 6 ' A detailed explanation will be given while explaining the following methods.

FIG. 9 is a flowchart sequentially showing the operation sequence of the image editing module according to the present invention, FIG. 10 is an image showing an image image processed by the image editing module, FIG. 11 is an outline FIG. 12 is a view showing a process of extracting a shaded part by the image editing module, FIG. 13 is a view schematically illustrating a contour shape comparison of a ground water image, FIG. 14 is a view An image showing an image processed by the image editing module will be described with reference to FIG.

S181; Steps to check the shooting point

In the video image processing system according to the present invention, the video image search unit 61 confirms the first ground image 20a shown in FIG. 10 (a), identifies the third ground image 20a as the first ground image 20a, It is necessary to search all of the image DB (1) in which the ground image 20a " is photographed (see the drawing of FIG. 10B) The image images including the same ground image as the first ground image 20a will be classified and managed together and the image image search unit 61 will display the GPS coordinates of the first ground image 20a And searches all of the image images having the GPS coordinates within a certain range in the image DB (1).

For reference, the video image search unit 61 transmits the GPS coordinate information of the search object to the image search module 2, and the image search module 2 searches the image DB 1 for the video image And transmits the retrieved image image to the image image retrieving unit 61. [0050]

S182; Ground water image contour confirmation step

The image outline confirmation unit 62 checks the corresponding image image in the image processing shown in FIG. 10A and the image image shown in FIG. 10B, Check the outline of

Here, the outline of the ground water image means the shape of the ground water image seen in the image. The outline of the ground water image is a known technique for tracking the boundary of the ground water image by separating the pixels on the basis of the color described in the image search module 5. The image outline confirmation unit 62 includes And then traces all the contours of the ground water images to complete the unique contour shape. Furthermore, the image outline confirmation unit 62 distinguishes the planar part and the side part from the outline shape of the ground water images, and extracts only the planar part of them, thereby completing the image shown in FIG. For reference, the division of the plane portion and the side portion is also based on color and contrast.

S183; Modified ground water image search

The ground water image confirmation unit 63 confirms the contour of the first ground image 20a in the first image (the drawing of FIG. 11 (a)) confirming the contour of the corresponding image image (the drawing of FIG. 10 A first contour shape and a second contour shape of another ground image adjacent to the first contour shape are confirmed and the arrangement state of the first and second contour shapes is confirmed. Next, the ground water image confirmation unit 63 confirms the contour shape of the ground water images in the second image (the view (b) of FIG. 11) that confirmed the outline of the other image (FIG. 10 Check the status.

Then, the ground water image confirmation unit 63 compares the first and second contour shapes of the first image and the contour shapes of the second image, and calculates a first ground image (" The third contour shape of the third terrestrial object image 20a "corresponding to the contour of the third contour shape 20a is confirmed. Further, the fourth contour shape which is another contour contour adjacent to the third contour shape is confirmed.

For reference, the ground-water image determining unit 63 searches the second image for a third outline shape that is the same as or similar to the first outline shape, and when an outline shape identical or similar to the first outline shape is identified, The outline shapes adjacent to the periphery are compared with the second outline shapes to determine whether the third outline shape and the fourth outline shape are formed.

On the other hand, even if the contour shapes shown in the first image and the second image are contours of the same ground-based image, different contour shapes can be obtained due to difference in shooting angle, and the intervals between the contour shapes can be different. Accordingly, the search for the third and fourth outline shapes corresponding to the first and second outline shapes is performed regardless of the interval, assuming that the same or similar outline shapes within the specified range form the same arrangement state.

In addition, although the first image and the second image are related to the image image of the same point, the outline shapes DS1 and DS2 of the specific ground image may not be confirmed by the shade. In this case, if the contour shapes DS1 and DS2 of the specific ground object image are less than the specified number, it is ignored, and if the number is greater than or equal to the specified number, the corresponding second image is regarded as an image at another point and is excluded.

S184; Same ground image image search step

Once the third and fourth contour shapes corresponding to the first and second contour shapes are identified, the ground water image verification unit 63 determines the first and third contour shapes, which correspond to the first and second contour shapes, A second zone formed by the boundary, and superimposes the centers of the first zone with the first profile and the second zone with the third profile as shown in FIG.

In this overlapping state, the position of the first contour shape and the position of the third contour shape are compared. The ground surface image confirmation unit 63 determines the distance (BD) between the plane portion of the first contour shape and the plane portion of the third contour shape, . The ground water image confirmation unit 63 compares the first contour shape and the third contour shape of the second images one by one in this manner to check the separation distance BD between the flat parts of the first and third contour shapes, The second image having the longest separation distance BD among the identified separation distances BD and the corresponding third contour shape are confirmed.

The reason why the first contour shape and the third contour shape having a long separation distance BD are searched is that the planar portion of the first contour shape due to the photographing angle covers the ground located opposite to each other as shown in Fig. The shaded portion D is formed so that the actual image of the shaded portion D of the first ground-based image 20a corresponding to the first contour shape is divided into the third ground-based image (FIG. 20a ").

S185; Steps for selecting the actual image range

The first contour shape and the third contour shape having the longest separation distance BD are confirmed, the shade part processing unit 64 determines that the third ground image 20a " corresponding to the third contour shape is the first ground image & The range of the actual image corresponding to the shaded portion D of the water image 20a is selected.

In the embodiment of the present invention, the range of the image of the rough image is set to a range in which the first contour shape and the third contour shape do not overlap each other. As a result, the actual image has a range from the plane portion of the first ground image 20a to the second side portion of the third ground image 20a ".

S186; Actual image editing steps

The shaded part processing unit 64 extracts the actual image by the range from the video image including the third ground image 20a ", and outputs the extracted resolution, size and color of the actual image to the first ground image 20a ), The actual image is edited such that it can be synthesized with the corresponding image image according to the resolution, size, and color of the image.

S187; Image image synthesis processing step

When the first ground image 20a is imaged to the second ground image 20a 'by the image processing module 7, the shaded part processing unit 64 processes the shaded part D generated in the image processing, The actual image is synthesized as shown in FIG.

As a result, the shaded portion (D), which was obscured by the ground water image, is removed, and the shaded portion (D) is automatically restored through the synthesis of the actual image, thus completing a neat image image without the portion covered by the ground water image .

S188; Edit placement location step

When the second ground image 20a 'is completed according to the above-mentioned procedure and the shaded portion D is synthesized so as to be visible, the second ground image 20a' is correctly placed in place .

To this end, as shown in FIG. 15 (a view showing a state prior to the modification of the arrangement point of the ground image) and FIG. 16 (the view after the arrangement point modification of the ground image) GPS1 'and the' GPS2 ', which are the GPS coordinates of the ground, which is the object of the' preimage image 'which is the second ground image 20a'. Here, the GPS coordinate system is formed in a lattice form according to a constant numerical value based on a reference point defined in the image. For reference, in the embodiment of the present invention, the GPS coordinates of the 'reference point' are (12.32, 21.72) and the GPS coordinate system is formed at intervals of 0.01.

The position correcting unit 65 confirms the GPS coordinates of the GPS coordinate system set in the corresponding image and the GPS coordinates of the 'corrected image' to be corrected. Here, the GPS coordinates 'GPS1' and 'GPS2' of the 'preimage image' are (12.40 and 21.745) and (12.42 and 21.76), respectively.

By comparing the GPS coordinate system with 'GPS1,2', it is confirmed that it is not located at the designated position. The position correcting unit 65 confirms this and moves the 'pre-corrected image' to the corresponding position in the GPS coordinate system according to 'GPS1,2' to complete 'corrected image'. At this time, the position correcting unit 65 also moves the corresponding shaded portion D of the second ground image 20a ', which is the target of the' before and after image ', at the time of the modification.

The position correcting unit 65 compares the hue and contrast of the second ground image 20a 'and the shaded portion D with the hue and the contrast of the surrounding image, respectively, The color and contrast of the ground-water image 20a 'and the shaded portion D coincide with the hue and contrast of the surrounding image.

In the case where the ground water image is modified according to the process described above, various shading sections included in the image image are removed, and the arrangement status of the ground water images posted on the image image can be easily grasped. In addition, The user can grasp and utilize the corresponding video image with high reliability.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (1)

An image DB 1 for storing image image data to which GPS coordinates are applied, image height h2 of the image image, and height information h1 of the ground image; An image retrieval module 2 for retrieving data and information of the image DB 1; An output module 3 for outputting image image data retrieved by the image retrieval module 2 through a touch screen 3a having an input function; (P1) and a pair of end points (P2) in a video image outputted to the touch screen (3a), and sets a reference straight line connecting the coordinate values to obtain an initial point P1 ) And the end point (P2) to the range of the road image (30); A predetermined color of a pixel corresponding to the road image 30 is confirmed and a boundary of the pixel range designated by the same color as the designated color of the road image 30 on the reference straight line is inconsistent, A1 to a6 of the first ground image 20a and the first ground image 20a forms a search straight line in a direction covering the road image 30 and outputs the same through the output module 3 It is possible to confirm the designated color of the pixel corresponding to the selected point on the search straight line and confirm the range of the same pixel as the boundary to the boundary of the reference ground image 10a, The reference point b1 to b4 of the reference ground image 10a and the width w2 of the first ground image 20a of the first ground image 20a through the reference points a1 to a6 of the first ground image 20a, An image search module 5 for calculating an image of an object; The center distance d between the reference ground 10 and the ground 20 and the height h2 of the ground and the photographing height h2 of the image are checked

(W2) and the photographing angle ([theta]) of the first ground-based image 20a are calculated

(L2) of the first ground-water image (20a), and adjusts the size of the first sub-surface image (20a) in accordance with the modified plane width (L2) An image processing module (7) for removing a side image of the ground image (20a) to complete a second ground image (20a '),
And a third ground image 20a " obtained by photographing the ground water of the first ground water image 20a from another angle, based on the GPS coordinates of the image image data including the first ground water image 20a A video image retrieval unit (61) for retrieving the video image data from the image DB (1);
The outline in the form of the first ground image 20a and the third ground image 20a " is checked based on the boundary of the ground image images confirmed by the image search module 5, and the first ground image 20a ) And the third terrestrial image 20a "in the outline of the first terrestrial image 20a ", respectively, and extracting only the planar portion and forming the outline of the first terrestrial image 20a, An image contour confirmation unit 62 for completing a second image having an outline of the image 20a ";
The first contour of the first ground image 20a is a first contour, and the contour of other ground images adjacent to the first contour is identified as a second contour, , And confirms the arrangement state of the two contour shapes. The third contour shape corresponding to the first contour shape and the third contour shape corresponding to the second contour shape correspond to the flat portion and the side portion of the third terrestrial object image 20a " And a second region formed by a boundary extending to the fourth outline shapes, respectively, and the first contour shape is determined by the first contour shape, And a third area having a third contour shape, and comparing the positions of the first contour shape and the third contour shape with each other to compare the positions of the first contour shape and the third contour shape with each other, (BD) between the flat part of the contour shape, Interval distance (BD) a second third ground water image check unit 63 to determine the outline form corresponding to the image that has the longest distance (BD) of the;
The range of the actual image corresponding to the shaded portion D is determined to be a range in which the first contour shape and the third contour shape do not overlap with each other, Size and color of the extracted actual image in accordance with the resolution, size, and color of the video image including the first ground image 20a, and outputs the edited actual image as an image A shaded portion processing unit 64 for synthesizing the shaded portion D to a shaded portion D of the video image including the two-ground-water image 20a ';
When the difference between the point where the second ground image 20a 'is located based on the GPS coordinate system linked to the video image and the GPS coordinates of the ground object which is the object of the second ground image 20a' The position of the second ground image 20a 'is corrected to one point of the GPS coordinate system corresponding to the GPS coordinates, and the color and the contrast of the second ground image 20a' A position correcting unit 65 for adjusting the position;
An image editing module 6,
And an image processing unit for processing the image of the point recognition and automatic correction method for updating a video image.

Images