KR101762613B1 - Panorame Image Greneration Method for Tube Hollow Photograph - Google Patents

Abstract

According to the present invention, provided is a method of generating a panoramic image of photographs (a) of a hollow tube inner surface (1), including an orthoscopic image (10) and a side-view image (20). The method comprises a first step (S100) of extracting only the orthoscopic image (10) from the photographs (a); a second step (S200) of generating image division information (b) for the orthoscopic image (10); and a third step (S300) of generating the panoramic image by connecting a plurality of individual images (30a, 30b, 30c) divided by the image division information (b). According to the present invention, it is possible to automatically classify the orthoscopic images (10) included in the existing photographs (a) and to easily generate the panoramic image of the hollow tube inner surface (1).

Description

[0001] The present invention relates to a method for generating a panoramic image of a hollow tube photographic image,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a construction field, and more particularly, to a method for generating a panoramic image of a hollow tube photographed image of an inner surface of a hollow pipe structure such as a sewage pipe.

In the case of a hollow pipe structure, it is necessary to check the inner surface condition for construction and design. In particular, in the case of sewer lines, it is required to check whether cracks have occurred for maintenance.

Such an inner surface condition check is performed through an image taken on the inner surface. The inner surface of the hollow tube is generally used with special equipment capable of rotating the camera 360 degrees or photographing 360 degrees.

However, in the case of the method of photographing the camera by rotating the camera 360 degrees, the photographing range is not constant, so that it is difficult to photograph the inner surface of the uniform quality and the time required for the photographing is long.

In addition, in the case of special equipment capable of shooting 360 degrees, it is required not only to construct expensive equipments but also to have an image editing technique of shooting 360 degrees.

In order to solve this problem, a panoramic image is generated through image processing of a photographed image of a sewer pipe to perform image reading.

The technique of generating a panoramic image has a method of using a fisheye lens using a fisheye lens capable of 360-degree wide angle photographing, but it is not well used because of high price of the apparatus and difficulty in management.

In addition, many existing sewerage exploration videos held by local governments are images taken using a general CCTV camera.

In order to observe the time-lapse images (Figs. 1 and 10) of the frontal direction of the sewer line and the damaged area of the inner surface of the sewer line, images taken using the CCTV camera were taken with the side- Are mixed together.

There is a problem in that it is necessary to separately extract and edit only the regular image 10 for generating a panoramic image because it takes a great deal of time and money to perform a separate photographing operation to generate a panoramic image.

The present invention is conceived to solve the problem of generating a panoramic image of the above-mentioned hollow tube photographing image. The object of the present invention is to automatically extract only a regular time image from a hollow tube photographing image in which a regular image and a time- And a method for generating a panoramic image of a hollow tube photographing image.

In addition, the present invention provides a method of generating a panoramic image of a hollow tube photographic image, which enables an accurate panoramic image to be generated using automatically extracted time-based images.

According to an aspect of the present invention, there is provided a method of generating a panoramic image of an image (a) of a hollow tube inner surface (1) including a regular image (10) and a temporal image (20) A first step S100 of extracting only the regular image 10; A second step (S200) of generating image classification information (b) for the regular image (10); And a third step (S300) of connecting the plurality of individual images (30a, 30b, 30c) partitioned by the image segmentation information (b) to generate the panorama image (S300) A panoramic image of the < / RTI >

In this case, the photographed image (a) further includes photographed time information (c), and the first step (SlOO) is a step of acquiring Cross Correlation Value (CCV) of the photographed image (a) Information deriving step (S110) of deriving information d; And a comparison step (S120) of comparing the CCV information (d) with a mask filter image (e). The panoramic image of the hollow tube photographed image may be generated.

In addition, the CCV information (d) may be derived by the following formula (1), and may be a method of generating a panoramic image of the hollow tube photographed image.

[Formula 1]

CCV (t): CCV information (d) at t (0 to 65535)

m x n: resolution (m pixels in the vertical direction, n pixels in the horizontal direction) of the photographed image (a)

t: recording time information (c)

: 촬영영상(a)의

행

열 픽셀의 시간

에서의 명도(0~255)

: Of the photographed image (a)

line

The time in column pixels

The brightness (0 to 255)

: 마스크 필터 영상(e)의

행

열 픽셀의 명도(0~255)

: Of the mask filter image (e)

line

Brightness of the column pixels (0 to 255)

When the CCV information d is equal to or greater than the reference value f, the CCV information d is discriminated to be the temporal image 20. If the CCV information d is less than the reference value f, The method of generating a panoramic image of a hollow tube photographed image.

In addition, the image segmentation information (b) includes a plurality of straight line information (b1) extending along the longitudinal direction of the hollow tube expressed in the regular image (10); And a plurality of annular line information (b2) connecting the inner circumferential surface of the hollow tube represented in the regular image (10).

The outline g of each of the plurality of individual images 30a, 30b and 30c partitioned by the image segmentation information b is formed by a pair of the annular shapes b1, And the line information b2 of the panoramic image of the hollow tube.

In the third step S300, the outline g is converted into a rectangle by changing a pair of the annular line information b2 included in the plurality of individual images 30a, 30b, (S310); And an arrangement step (S320) of arranging the plurality of individual images (30a, 30b, 30c) changed in the rectangle to the contour g in the order of the neighboring individual images. It may be a method of generating a panoramic image.

In addition, the photographed image (a) further includes photographing distance information (h), and the third step (S300) And a correction step (S330) of correcting an arrangement error of the image. The panoramic image of the hollow tube photographed image may be generated.

According to another aspect of the present invention, there is provided a computer-readable recording medium on which a program for executing a method of generating a panoramic image of a hollow tube photographing image is recorded.

According to the present invention, there is an effect that only a regular time image can be automatically extracted from a hollow tube photographed image in which a regular image and a side view image are mixed.

According to the present invention, an accurate panoramic image can be generated using the automatically extracted time-lapse image.

FIG. 1 is a view showing a regular image according to an embodiment of the present invention. FIG.
2 is a view showing a side view image according to an embodiment of the present invention;
3 is a view showing a mask filter image according to an embodiment of the present invention.
4 is a diagram illustrating a comparison step of comparing a temporal view image and a mask filter image according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating a comparison step of comparing a regular image and a mask filter image according to an embodiment of the present invention. FIG.
6 is a view showing an image at the time of addition of image classification information according to an embodiment of the present invention;
FIG. 7 is a diagram illustrating a conversion step of converting an outline of an individual image into a rectangle according to an embodiment of the present invention; FIG.
8 is a diagram illustrating an arrangement step of arranging a plurality of individual images whose outlines are transformed into rectangles according to an embodiment of the present invention.
9 illustrates a panoramic image generated in accordance with an embodiment of the present invention.
10 is a diagram illustrating steps of a method of generating a panoramic image of a hollow tube photographic image according to an embodiment of the present invention.
11 is a diagram showing a configuration of an apparatus in which a method of generating a panoramic image of an image of a stern tube tube is performed in an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a perspective view of a panoramic image of a hollow tube photographed image according to the present invention; Fig. And redundant explanations thereof will be omitted.

It is also to be understood that the terms first, second, etc. used hereinafter are merely reference numerals for distinguishing between identical or corresponding components, and the same or corresponding components are defined by terms such as first, second, no.

In addition, the term " coupled " is used not only in the case of direct physical contact between the respective constituent elements in the contact relation between the constituent elements, but also means that other constituent elements are interposed between the constituent elements, Use them as a concept to cover each contact.

The present invention relates to a method of generating a panoramic image using a photographed image (a) of a previously captured hollow tube inner surface (1).

The most representative hollow tube equipped with the photographed image (a) has a sewer line.

The local governments responsible for the maintenance and management of the sewerage line shoot the sewerage line by CCTV and read the generated image (a) to search for problems such as cracks inside the sewerage line.

The photographed image (a) includes not only the regular image 10 of the front of the sewage pipe but also a side view image 20 of a specific side wall. The regular image 10, which is the basic data for generating the panoramic image, it is required to be classified in (a).

However, since the frequency and position of the regular image 10 and the side view image 20 are not regular, a large amount of work is required to identify and classify the entire captured image a.

In order to solve such a problem, the present invention proposes a method of generating a panoramic image of a hollow tube photographic image which can automatically read and classify the regular image 10 in the photographed image (a).

Furthermore, a method for generating a panoramic image of a hollow tube photographic image for generating a panoramic image of the hollow tube inner surface 1 based on the classified temporal image 10 is presented.

Hereinafter, a method of generating a panoramic image of a hollow tube photographic image according to an embodiment of the present invention will be described with reference to the accompanying drawings.

A method of generating a panoramic image of a hollow tube photographic image according to an exemplary embodiment of the present invention includes a first step S100 of extracting only the regular image 10 from the photographed image a, A third step S300 of generating a panoramic image by connecting a plurality of individual images 30a, 30b and 30c partitioned by the second step S200 of generating the information b and the image classification information b, .

The first step S100 is a step of reading and extracting the regular image 10 existing in the photographed image a.

The reading and extraction of the regular time image 10 uses the CCV (Cross Correlation Value) information d of the photographed image corresponding to the photographing time information c.

Cross Correlation Value (CCV) information d is derived numerically by comparing the brightness of the photographed image (a) with the brightness of the mask filter image (e).

The mask filter image e is a reference devised for reading the regular image 10, and is an image expressed by abstracting the regular image 10 with a brightness that is common to the regular image 10 (FIG. 3).

Therefore, the brightness of the mask filter image e and the photographed image a is compared to determine the image that does not make a large difference as the regular image 10, and the extracted image is automatically extracted.

In order to automatically perform the extraction process, information that numerically expresses the difference between the mask filter image (e) and the photographed image (a) is defined as CCV (Cross Correlation Value) information (d).

If the cross correlation value (CCV) information d is equal to or greater than the preset reference value f, it means that the difference in brightness of the two images is large. If the value is less than the reference value f, it means that the difference in brightness of the two images is small .

Thus, if the CCV information d is equal to or greater than the reference value f, it is determined to be the temporal image 20, and if the CCV information d is less than the reference value f, it can be determined to be the regular video 10.

Such discrimination can be performed by the control unit 100 in an apparatus such as a computer or an information communication apparatus in which the control unit exists, and the results of the series of processes can be displayed to the reader through the display unit 200, 300).

On the sewerage inspection image taken by using the existing sewer pipe exploration device, the time image of the camera facing the front is different from the state of the pipe and the operation of the investigator, but it is generally about 20 to 30% of the total time of the image.

If CCV (t) is changed according to time t, CCV will be 0 ~ 65535.

The interval that shows the value of the lower 20% (empirical value) in the CCV over the entire interval can be discriminated from the regular image.

Specifically, the photographed image (a) includes the photographed time information (c) at which the image was photographed.

-> Explanation of CCV (Cross Correlation Value) concept of photographed image is required.

의 명도 분포가 마스크필터

의 명도분포와 얼마나 다른 지를 매 시간마다 수치적으로 나타내는 함수입니다. 만약 마스크 필터가 전부 검은색(

(

,

)=0)이고 특정 시간

의 조사영상이 전부 흰색(

)인 경우 CCV는 65535이고 마스크필터와 동일하게 조사영상이 전부 흰색인 경우 CCV는 0이 됩니다. 만약 시간

에서 조사영상이 마스크필터와 동일하다면(

) CCV는 0이고, 마스크 필터의 명도 분포와 달라질수록 큰 CCV를 나타내게 된다. The CCV is defined as the survey image

Lt; RTI ID = 0.0 >

This is a function that numerically displays how much the brightness distribution differs from the brightness distribution at every hour. If the mask filter is all black (

(

,

) = 0) and a specific time

Of the surveyed images are all white (

), The CCV is 65535 and CCV is 0 when the irradiation image is all white as the mask filter. If time

If the illuminated image is the same as the mask filter (

) CCV is 0, and the larger the chromaticity distribution of the mask filter is, the larger the CCV is.

Therefore, when the average brightness distribution of the regular image is used as the mask filter, it shows the CCV when it is the regular image or the CCV which is similar to the regular image. Thus, it is possible to discriminate the regular video.

에서 촬영영상(a)의

행

열 픽셀의 시간

에서의 명도를

로 정의하고 마스크 필터의 명도를

로 정의하는 경우 CCV(Cross Correlation Value) 정보(d)는 다음과 같이 정의된다. 명도 값은 0~255(검정: 0, 희색: 255)의 범위에 있다. When the photographed image (a) has the resolution of "n x m ", the photographed time information (c)

Of the photographed image (a)

line

The time in column pixels

The brightness in

And the brightness of the mask filter

The Cross Correlation Value (CCV) information (d) is defined as follows. Brightness values range from 0 to 255 (black: 0, white: 255).

This can be expressed as [Equation 1] below.

[Formula 1]

CCV (t): CCV value at t (d) (0 to 65535)

m x n: resolution (m pixels in the vertical direction, n pixels in the horizontal direction) of the photographed image (a)

t: recording time information (c)

: 촬영영상(a)의

행

열 픽셀의 시간

에서의 명도(0~255)

: Of the photographed image (a)

line

The time in column pixels

The brightness (0 to 255)

: 마스크 필터 영상(e)의

행

열 픽셀의 명도(0~255)

: Of the mask filter image (e)

line

Brightness of the column pixels (0 to 255)

와

에

와

가 각각 1~

, 1~

까지 순차적으로 대입되는 형태이다.
In the CCV (t) estimation equation, each function

Wow

on

Wow

Respectively.

, 1 ~

In order.

The first step S100 according to the embodiment of the present invention derives information for deriving CCV (Cross Correlation Value) information d of the photographed image a corresponding to the photographing time information c, (S120) comparing the CCV information (d) with the mask filter image (e).

When the first step S100 is performed, only the regular image 10 is automatically extracted from the photographed image a.

The second step S200 is a step of generating the image segmentation information b for dividing the regular image 10 into the individual images 30a, 30b, 30c, and the like for panoramic image formation.

The outline g of the plurality of individual images 30a, 30b and 30c partitioned by the image segmentation information b is divided into a pair of annular line information b2 facing the pair of facing straight line information b1 (Fig. 7).

The straight line information b1 extends along the longitudinal direction of the hollow tube represented in the regular image 10 and is expressed in plural along the circumference of the hollow tube, (Fig. 6).

The annular line information b2 is represented by an annular line connecting the inner circumferential surface of the hollow tube represented in the regular image 10 and is expressed in plural along the longitudinal direction of the hollow tube (Fig. 6).

In a third step S300, the outline g of the individual image 30a is converted into a straight line, and neighboring individual images are sequentially arranged and connected to generate a panoramic image.

Since only the portion represented by the annular line information b2 out of the outline g of the individual images 30a, 30b, 30c and the like is formed as a curve, the annular line information b2 is changed to a straight line, ) Are rearranged in a straight line according to the change of the outline g.

Then, a plurality of individual images in which all the outlines g are formed in a straight line are arranged and connected in the order of neighboring individual images.

Accordingly, in the third step S300 according to the embodiment of the present invention, the pair of annular line information b2 included in the plurality of individual images 30a, 30b, and 30c is linearly changed to form the outline g A transformation step S310 of transforming the rectangle into a rectangle and an arranging step S320 of arranging the plurality of individual images 30a, 30b and 30c changed into rectangles in the order of neighboring individual images.

The third step S300 according to an embodiment of the present invention includes a correction step of correcting an arrangement error of individual images arranged in the arrangement step S320 using the shooting distance information h included in the shooting image a, (S330).

The sewer pipe exploration device has a sensor that can measure the penetration distance of the device into the sewer pipe in the cable winder, and is displayed on the exploration image. Even when the image is converted using the CCV, the area A3x close to the camera initially has a high resolution and the area A1x far from the camera has a relatively large error.

It is possible to reduce the error by converting the image for each entering distance and converting it into a method for eliminating the image having a large error in the distant area.

A method of generating a panoramic image of a hollow tube photographic image according to an exemplary embodiment of the present invention may be implemented in the form of a program command that can be executed through various computer means and recorded on a computer readable medium.

The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like.

Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention as defined in the appended claims. It is to be understood that both the technical idea and the technical spirit of the invention are included in the scope of the present invention.

100: camera
200: pointer forming member
300:
400:
500:
600: Monitoring section

Claims (9)

1. A method for generating a panoramic image of a captured image (a) on an inner surface (1) of a hollow tube including a regular image (10) and a temporal image (20)
A first step (S100) of extracting only the regular image (10) from the photographed image (a);
A second step (S200) of generating image classification information (b) for the regular image (10); And
And a third step (S300) of connecting the plurality of individual images (30a, 30b, 30c) partitioned by the image segmentation information (b) to generate the panorama image,
The photographed image (a) further includes photographing time information (c)
The first step S100 may include an information derivation step (S110) of deriving CCV (Cross Correlation Value) information (d) of the photographed image (a) corresponding to the photographing time information (c) And
(S120) for comparing the CCV information (d) with the mask filter image (e)
And generating a panoramic image of the hollow tube photographed image. delete The method according to claim 1,
Wherein the CCV information (d) is derived by the following formula (1).

[Formula 1]

CCV (t): CCV value at t (d) (0 to 65535)
m x n: resolution (m pixels in the vertical direction, n pixels in the horizontal direction) of the photographed image (a)
t: recording time information (c)

: Of the photographed image (a)

line

The time in column pixels

The brightness (0 to 255)

: Of the mask filter image (e)

line

Brightness of the column pixels (0 to 255)
The method of claim 3,
If the CCV information d is equal to or greater than the reference value f, it is determined that the CCV information d is the time-
And when the CCV information (d) is less than the reference value (f), it is determined to be the regular image (10). 5. The method of claim 4,
The image segmentation information (b)
A plurality of straight line information (b1) extending along a longitudinal direction of the hollow tube expressed in the regular image (10); And
A plurality of annular line information (b2) connecting the inner circumferential surface of the hollow tube represented in the regular image (10);
And generating a panoramic image of the hollow tube photographed image.
6. The method of claim 5,
The outline g of each of the plurality of individual images 30a, 30b and 30c partitioned by the image segmentation information b is formed by a pair of the straight line information b1 facing each other, (b2). < Desc / Clms Page number 24 >
The method according to claim 6,
In the third step S300,
A converting step (S310) of converting a pair of the annular line information (b2) included in the plurality of individual images (30a, 30b, 30c) into a rectilinear shape and converting the outline line (g) into a rectangle; And
(S320) of arranging the plurality of individual images (30a, 30b, 30c) having the rectangle of the outline (g) in the order of the neighboring individual images;
And generating a panoramic image of the hollow tube photographed image.
8. The method of claim 7,
The photographed image (a) further includes photographing distance information (h)
In the third step S300,
A correction step (S330) of correcting an arrangement error of the individual images arranged in the arrangement step (S320) using the photographing distance information (h)
Wherein the panoramic image of the hollow tube photographed image is generated.
A computer-readable recording medium having recorded thereon a program for executing a method of generating a panoramic image of a hollow tube photographic image according to any one of claims 1 to 8.

Images