KR20090078447A - The measuring method for distance and area by digital cameras, and the calcurating method for damage information by the same measuring method, and the calcurating system for damage information which is used in the same calcurating method - Google Patents

Abstract

A method for measuring distance and surface through a digital camera and a method for calculating disaster information and a disaster information calculating system are provided to calculate disaster information produced by the decimal manpower by calculating the disaster information. A horizontal bar(4) is formed longer than into a traverse direction. Two cameras are separated at constant interval in an upper part of the horizontal bar. The cameras are arranged in order to face the respective front side. A support unit(3) supports arm from the surface. A computer stores and receives image information which the cameras take a picture. The computer computes a range and damage compensation amount of money of the damaged area on basis of a program.

Description

The measuring method for distance and area by digital cameras, and the calcurating method for damage information by a method of measuring distance and area using a digital camera, a method of calculating damage information using the same, and a damage information calculating system used therein the same measuring method, and the calcurating system for damage information which is used in the same calcurating method}

The present invention relates to a method for measuring distance and area with a digital camera, a method for calculating damage by simply and accurately measuring a damage range of a disaster area using the same, and a damage information calculation system used therein. .

That is, in the present invention, it is possible to convert the image data photographed using two digital cameras to three-dimensional data through the improvement of the image processing technique and the three-dimensional coordinate reading method, through which the distance and The area can be calculated relatively accurately.

Furthermore, in the present invention, the method of measuring the distance and the area with a digital camera is applied to calculate the damage range of the disaster area, so that the disaster is much more accurate and convincing than the conventional method depending on the neck. Provide information (distance and area of the affected area).

Frequent weather events since 1998 have been recognized as new trends that are no longer accepted as extreme weather events, and the feng shui damage caused by such weather changes frequently results in frequent damages and also results in enormous damage. As a result, it is emerging as a major concern of national capacity that can no longer be observed.

At present, domestic disaster prevention system is established as a response system after disaster occurrence, rather than focusing on the prevention system for disaster preparedness, and even the investigation for disaster recovery does not have a uniform survey system.

The current damage investigation system for damage caused by various feng shui damages is a fundamental problem such as the lack of disaster prevention system such as the lack of time and the number of investigators, additional expenditure of enormous national budget, and the possibility of national waste due to inaccurate estimates. Since it is difficult to investigate damages through huge replenishment, a unified damage investigation system for various types of damages is urgently needed.

In addition, damage investigation and estimation of damage estimate due to frequently occurring feng shui damage may lead to incorrect results due to information distortion, and therefore, it is required to calculate damage estimate quickly and accurately by investigating the damage area in a rational and scientific way. In order to rationalize early disaster recovery, it is necessary to develop a system for quickly and accurately counting damage estimates and identifying reasonable damages.

The scope of damage investigations on public facilities, which are currently mainly carried out in the country, is generally surveys of local or local damage sites. In the investigation of the local damage site, the damage range is directly observed on the site only by disaster prevention personnel, so the grounds for objective damage calculation are insufficient and the calculation of the damage is difficult, and the maintenance of damage information is also difficult.

Therefore, some systems and techniques for acquiring disaster information using digital damage image data from local damaged areas and systems for estimating the damage estimation amount of damaged areas have been developed for public facilities. It is a stage where technology to acquire disaster information by using video data and to easily estimate damage estimate is developed.

However, it is necessary to improve the processing speed and transmission technology of on-site damage video, and to improve the module for estimating damage estimation. There is an urgent need for technical improvement.

Therefore, this study aims to improve the technology for objective and reliable disaster information and estimation of damage estimates by standardizing disaster scene image information system and acquiring more accurate disaster information from local damage area of public facilities.

The present invention has been made to solve the above problems,

In the past, the method of measuring the disaster range (distance and area of the disaster area), which mainly depended on the neck or tape measure, can be easily calculated using image data from two digital cameras. The first purpose is to make it easy to calculate.

In addition, the two-dimensional image data provided by the two digital cameras is first converted into three-dimensional data by the improved image processing method and the three-dimensional coordinate reading method provided by the present invention, and then the disaster area by this value. By calculating the distance and area of the second object is to be able to calculate the data that is not much different from the actual by only photographing.

Finally, the third purpose is to make the equipment for calculating disaster information to be in a compact and stable state so that it can be used conveniently even if it is not a professional man.

In the present invention,

Two cameras, a computer to which two-dimensional image data captured by the camera are transmitted, and damage quantity calculation for calculating distance and area between coordinates by converting two-dimensional image data into three-dimensional data in the computer. By the program and the damage estimating program for calculating the damage estimating amount by the calculated damage amount (distance or area), the objects of the present invention are achieved.

That is, the shooting system, damage quantity and damage estimation program are linked to one system, so damage situation and cause analysis, damage quantity calculation, damage estimation amount calculation, etc. are done at the damage site. The affected area can be investigated.

On the other hand, in the digital camera, a geometrically faithful image can be obtained in real time without a separate storing and processing process by using an optoelectronic sensor and an array of image sensor (CCD) array sensors of various sizes. The image is formed on an image pickup device such as a CCD or a CMOS instead of a film, and the mosaic-shaped image information obtained here is measured by the number of pixels, but the image information captured by two cameras is partially overlapped. By calculating the distance from, we can convert 2D image data into 3D data.

)를 결정한 후에야 비로소 2차원 영상 데이터를 3차원 데이터로 변환하여 각 좌표간의 거리 또는 면적을 산출해내게되므로, 사진 촬영만으로도 실제값에 가까운 데이터를 얻을 수 있도록 한다.In particular, in the present invention, first, the internal expression element (the amount of movement of the principal point of the camera, the focal length, the lens distortion coefficient) suitable for the camera characteristic is determined, and then the external expression element (

) Is determined before the two-dimensional image data is converted into three-dimensional data to calculate the distance or area between the coordinates.

In addition, the damage information calculation system according to the present invention includes two cameras, a program for calculating the distance or the area between the coordinates from the image information by the camera, and the damage amount or the like from the distance or the area between the coordinates. It consists of a program, a computer including and executing the programs, and a support for stably supporting the two cameras so as not to shake during shooting. The configuration of the system is very simple and the overall weight of the equipment is light. Therefore, not only the mobility of the equipment is good but also has the advantage that does not require a large number of people to work.

In the past, the extent of the disaster (distance and area of the disaster area) was mainly measured by a measuring method using a tape measure or a tape measure. However, in the present invention, the measurement of the damage range and calculation of compensation amount are only performed using image data photographed with a digital camera. Since it is possible, not only can drastically reduce the number of working people, but also enables a faster work.

In addition, two-dimensional image data provided by two digital cameras is first converted into three-dimensional data by the improved image processing method and three-dimensional coordinate reading method provided by the present invention, and then the value of By calculating the distance and area, highly accurate data can be obtained.

Finally, the damage information calculation system according to the present invention is composed of only two cameras, a computer containing various programs, and a support for stably supporting the two cameras so as not to shake during shooting. Because of its compactness and compactness, it has good mobility.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.

1 is an exploded perspective view and a photograph of a damage information calculation system according to the present invention, FIG. 2 is a photograph of an indoor model board for internal expression in the present invention, and FIG. 3 is a monitor screen of a damage scale precision estimation program according to the present invention. Figure 4 is a flow chart of the damage scale precision calculation program operating procedure according to the present invention, Figure 5 is a view schematically illustrating the damage calculation process according to the present invention.

The damage investigation calculation method proposed by the present invention is a method of calculating damage quantity by a damage investigator of a first line by measuring a damage spot using a digital image processing technique and a camera.

This method has the advantage that the damage amount can be aggregated by calculating the damage quantity quickly and accurately, and the damage investigation equipment has the advantage of performing the damage investigation at a relatively low cost.

In the case of local damage of public facilities, most damages can be calculated through length measurement and digital image processing. Among them, river embankments and facilities cannot be directly accessed due to flood damage. Judging.

If direct length measurement is not possible due to the location, size, weather conditions, and accessibility of the affected area, it is relatively easy to calculate the damage quantity by using remote imaging system from the field. It would be a good alternative.

Therefore, the damage investigation calculation method according to the present invention is considered to be very applicable in the local damage investigation.

   In addition, the digital image processing technique can be applied to various facilities in the investigation of damages limited to public facilities, and the damage can be quickly acquired at the damage site by acquiring the image data at the site and calculating the damage quantity and the damage estimate. There is this.

Acquisition of damage images does not require specialized knowledge, so it is expected that non-experts can easily carry out damage investigation.

As the effectiveness of photogrammetry increases, research on digital photogrammetry, which acquires positional information on an object by processing digitized images, has been actively conducted. Various devices for acquiring digitized images have been developed and are widely used in general. Among them, digital cameras are relatively inexpensive devices and can easily acquire numerical images, and storage media are memory cards, hard disks, CD-ROMs, etc. Since various devices are available, there are many advantages in storing and reproducing data.

   The process of acquiring an image includes a camera capable of directly obtaining a digitized image and a method of acquiring a digital image by using a scanner for negative or positive print after printing using a general camera.

However, in the field of real-time photogrammetry or robot vision, digital images are acquired directly by using a digital camera. However, digital images are generally obtained indirectly by scanning a photo.

In this case, the error between the photographed image and the numerical image includes the printing process and the scanning process, and the image processing process also includes the error.

   Recently, high quality geometrically faithful images can be obtained in real time without separate storage and processing by photoelectric sensors and array of image sensor (CCD) array sensors. Its importance is highlighted in each field that requires observation, such as medicine, civil engineering, and architecture.

In the case of general survey photos, expensive coordinate observers should be used, but in the case of digital images, the position of the points in the image can be observed based on the number of matrixes of the image array.

Various devices for acquiring digitized images have been developed and are widely used in general. Among them, digital cameras are relatively inexpensive devices and can easily acquire numerical images, and storage media are memory cards, hard disks, CD-ROMs, etc. Since various devices are available, there are many advantages in storing and reproducing data.

In short, a digital camera can be defined as a camera that does not require film. It is a next-generation imaging device that captures an image through an imaging device (CCD) and stores it in a semiconductor instead of a film, which is a conventional method, so that the screen can be freely changed or reproduced through a PC or a printer without undergoing film development and printing.

   Like a general film camera, a digital camera performs the same optical functions as a film camera except for an image processing part. The image produced by the light incident through the lens is condensed on the image pickup device such as CCD or CMOS instead of the film, and then the mosaic image information is sent to the image processor. Create a computer file that can be viewed or printed directly on photo paper using a photo printer.

The difference in resolution of the digital camera is considered to affect the accuracy to some extent. The difference of one pixel in the image is not large at close range, but the difference is very large when the shooting distance is far. Therefore, it is advantageous to acquire the image by using a high resolution camera having excellent clarity and color.

On the other hand, in order to acquire the three-dimensional coordinate information of the object using a digital image acquired from a photographing system manufactured using two digital cameras, the facial expression process is required, which corresponds to the position of the projection center and the camera shooting direction for the two cameras. It is possible to extract three-dimensional coordinates of an object without reference point information only if the three-dimensional angle is determined before shooting.

The main purpose of the damage information calculation system according to the present invention is to extract three-dimensional information on the damaged object in the object space from two two-dimensional images. In order to acquire 3D information more accurately and reliably through the image capturing system, a geometric model suitable for camera characteristics should be applied. In order to establish a geometric model between the photographed image and the target space, an internal expression element suitable for the characteristics of the camera must be determined, and the internal expression element can be represented by a focal length, a position of a main point, and an image distortion coefficient.

   In order to determine the internal expression elements expressing the unique characteristics of each camera, various reference points may be installed in the object space and the method of obtaining them from the geometric model of the photograph may be applied. Since commercial digital cameras are not designed for surveying, the exact focal length and location of the pub are unknown and must be obtained from a precisely measured reference point.

   In order to obtain the internal and external expression elements unique to the digital camera by the direct linear transformation method, no photo coordinates or focal lengths are required, and the coordinates of the image and the three-dimensional coordinates of the object space are linearly related. It has the advantage of being expressed.

And compared to the Bundle method, it is not necessary to perform the calculation process of facial expression elements by inputting initial values and iterative calculation. However, in the case of direct linear transformation, the accuracy of the target coordinates is somewhat reduced, so the self-calibration bundle method using geometric model is applied to improve the accuracy.

   The effects of image distortion among the camera's internal components can be represented by radiation distortion, eccentric distortion, and atmospheric refraction, and can be mathematically modeled as distortion correction coefficients. Radiation distortion occurs in the process of changing the path of travel as the ray passes through the projection center of the lens and along the radial direction from the main point. The coefficient for correcting the distortion in the radial direction can be expressed as Equation (1-1) below.

(1.1)

(1.1)

는 방사방향 왜곡에 대한 계수이고 r은 다음과 같이 표현되는 주점에서부터의 방사거리(radial distance)이다. From here,

Is the coefficient for radial distortion and r is the radial distance from the principal point as

(1.2)

(1.2)

,

)는 임의의 상점에 대한 사진좌표이고, (

,

)는 주점의 사진좌표이다. 광각이 좁은 카메라나 정확도의 요구가 크지 않은 경우

항 하나만을 고려하여도 되며

항은 보다 높은 정확도가 요구되거나 카메라의 렌즈가 광각인 경우 적용할 수 있다.At this time, (

,

) Is the photo coordinate for any store, and (

,

) Is the photo coordinate of the tavern. Narrow wide angle cameras or accuracy requirements are not high

You may consider only one term

The term can be applied when higher accuracy is required or when the lens of the camera is wide angle.

   Decentering distortion is generally less than radial distortion and asymmetrical about the principal point. The eccentric distortion component in the x and y directions is given by Equation (1.3).

(1.3)

(1.3)

In this case, P1 and P2 are coefficients of eccentric distortion of the combined lens at infinite focal length. In most near field surveying, only P ₁ and P ₂ are considered. The effect of the improvement is minimal.

   The distortion correction factor combining the radial and eccentric distortions is given by Equation (1.4).

(1.4)

(1.4)

   The process of camera expression requires a model with accurate three-dimensional coordinates and sufficient target points must be distributed in space.

The extended collinear condition equation including distortion correction coefficients is shown in Equation (1.5), which means the mathematical relationship between the point in the object space and the point in the image.

(1.5)

(1.5)

: 영상좌표,

: 주점의 위치,

: 초점거리,

: 회전행 렬,

: 대상물좌표,

: 렌즈 투영중심의 위치이다. here,

: Image coordinate,

: Location of pub,

: Focal length,

: Rotation matrix,

: Object coordinates,

: Position of lens projection center.

Here, the position and focal length of the tavern are applied as a new unknown variable and six external expression elements are applied independently. The number of unknown variables is six external expression elements, two positions of the tavern, and one focal length. Nine, and can range from at least two to as many as six, depending on how the camera distortion coefficient is applied.

Equation (1.5) is nonlinear, so if we linearize it using Taylor series to calculate the internal and external expression elements of the camera, we can get the equations (1.6) and (1.7).

(1.6)

(1.6)

(1.7)

(1.7)

는 근사값

를 함수

에 대입하여 얻은 값,

는 보정량,

는 잔차이다. From here,

Is an approximation

Function

The value obtained by substituting for

Is the correction amount,

Is the residual.

If equations (1.6) and (1.7) are expressed as matrices, they are as shown in equation (1.8).

(1.8)

(1.8)

   In Equation (1.8), each matrix can be simply expressed as Equation (1.9).

(1.9)

(1.9)

   If only some of the elements of the B coefficient matrix are expressed in Eq. (1.9), then for externally expressed elements, Eq. (1.10) is used.

(1.10)

(1.10)

here,

.

.

과

를 찾았을 경우, 외부표정요소를 변수로 하는 선형방정식을 형성하여 삼차원 위치를 결정할 수 있다.When the internal expression element including the external expression element and distortion correction coefficient of the camera is determined, the object coordinate in the object space can be calculated from the point on the photograph using the extended collinear condition. Same point in left and right pictures

and

If we find, we can determine the three-dimensional position by forming a linear equation with the external expression element as a variable.

,

점에 대해 식(1.11), 식(1.12)와 같은 공선조건식을 구성할 수 있다. 이때,

영상좌표는 주점의 위치(

)와 렌즈왜곡계수(

)가 고려된 좌표이다.

,

For the point, we can construct collinear conditional expressions such as equations (1.11) and (1.12). At this time,

The image coordinate is the location of the tavern (

) And lens distortion coefficient (

) Is considered coordinates.

(1.11a)

(1.11a)

(1.11b)

(1.11b)

(1.12a)

(1.12a)

(1.12b)

(1.12b)

   When Equations (1.11) and (1.12) are known, the ground reference point coordinates P (X, Y, Z) can be calculated immediately. For each of X, Y, and Z,

(1.13a)

(1.13a)

(1.13b)

(1.13b)

(1.14a)

(1.14a)

(1.14b)

(1.14b)

Same as Equations (1.13) and (1.14) can be simplified again using Equations (1.15) and (1.16).

(1.15a)

(1.15a)

(1.15b)

(1.15b)

(1.16c)

(1.16c)

(1.16d)

(1.16d)

At this time,

to be. If the above expression is expressed in matrix form, it is the same as Equation (1.17).

(1.17)

(1.17)

   Simply put, we can write this as (1.18)

(1.18)

(1.18)

는 식 (1.19)와 같이 구해진다.Solution by least squares

Is given by equation (1.19).

(1.19)

(1.19)

   In this solution, the collinear conditional equation is expanded by Taylor series, and the initial values of X, Y, and Z are substituted, and the result is almost identical to the solution obtained. It is difficult to estimate the initial value of the distance direction when the object is observed from the ground. Can be effectively used.

   Based on the mathematical model described above, the program was coded and applied to a digital camera. The applied model is the Sony R1 model and the applied model is the same as that of FIG. 2, and 81 rods are vertically mounted on a grid of 10 cm intervals. The coordinates of the reference point acquired using two deodorites are shown in the table below.

Station number X Y Z Station number X Y Z One 2464.88 559.27 8515.75 61 3514.33 -339.65 8530.5 2 2611.59 560.6 8415.73 62 3514.33 -339.65 8530.5 3 2761.07 558.31 8517.36 63 3661.46 -338.59 8332.26 4 2910.73 552.31 8217.74 64 2467.45 -488.37 8520.01 5 3063.07 557.16 8320.19 65 2612.94 -481.21 8320.23 6 3212.95 558.26 8420.62 66 2757.31 -486.43 8220.19 7 3360.9 556.79 8523.05 67 2913.6 -489.08 8522.06 8 3513.05 558.34 8525.06 68 3062.66 -487.75 8422.21 9 3661.63 558.06 8425.09 69 3213.75 -482.58 8323.83 10 2462.77 410.69 8419 70 3367.25 -481.52 8224.4 11 2614.18 405.37 8320.66 71 3515.46 -486.93 8426.34 12 2761.26 409.48 8419.73 72 3656.9 -491.57 8529.9 13 2914.13 407.23 8321.46 73 2459.18 -628.27 8412.76 14 3055.41 397.5 8223.08 74 2615.51 -634.99 8514.33 15 3212.1 411.69 8524.95 75 2764.76 -630.96 8413.89 16 3363.31 409.53 8425.44 76 2917.5 -627.53 8317.51 17 3509.69 400.98 8227.81 77 3066.64 -623.87 8218.49 18 3656.4 410.81 8529.3 78 3215.92 -632.11 8417.98 19 2465.72 265.23 8521.76 79 3364.68 -637.02 8520.53 20 2613.51 256.48 8222.24 80 3519.23 -629.38 8321.61

Model reference point coordinates

A total of 81 reference points were used to calculate the position lens distortion coefficients of the external expression element and focal length main point of the collinear condition equation. In this case, the eccentric distortion coefficient is omitted in the lens distortion coefficient, and in the radial direction, the radial distortion in the horizontal direction and the vertical direction are different.

parameter value Xo 3092.6 mm Yo -77.6 mm Zo 7284.6 mm o 0.035422 rad p -0.020732 rad k 0.014048 rad xp 0.051139 mm yp -0.074984 mm f 14.449 mm xk1 0.00046953 xk2 0.00000003 xk3 -0.00000001 yk1 0.00046171 yk2 0.00000009 yk3 -0.00000001

* Calculated camera internal and external expression factors and distortion coefficients

   In this case, the root mean square residual for each image coordinate was 0.404 pixels in the x-coordinate direction and 0.384 pixels in the y-coordinate direction.

   This imaging system takes about 80 ~ 90% redundancy and general image matching considers the entire image as a search area to find the conjugated image. However, since the calculation time becomes very long, the search area needs to be limited to a certain area near the conjugate point.

   An important limitation on the conjugate point position for finding the conjugate point more accurately and quickly in image registration is the epipolar line.

과 지상의 점 P에 의해서 정의된다. 에피폴라 선은 영상평면과 에피폴라 평면의 교차면이다. 따라서 공액점은 에피폴라 선상에 반드시 위치한다. 이러한 이유로 에피폴라 선은 영상 정합시 탐색영역을 상당히 감소시킬 수 있다. 에피폴라 선을 계산하기 위해서는 입체영상의 표정요소를 미리 알고 있어야 한다.The figure above shows the epipolar geometry. Epipolar plane is projected

And point P on the ground. The epipolar line is the intersection of the image plane and the epipolar plane. Therefore, the conjugate point must be located on the epipolar line. For this reason, the epipolar line can significantly reduce the search area during image registration. In order to calculate the epipolar line, it is necessary to know the facial expressions of the stereoscopic image.

   An image rearranged such that each row of images in the numerical image is parallel to the epipolar line is called an epipolar image. When image registration is performed using such an image, the search region for the conjugate point is limited to the same row in another image. That doesn't mean you have to search the entire epipolar line. By predicting the conjugate point position, the search area can be further reduced.

는 대상물의 예상 촬영거리를 나타내며 일반적으로 대상 지역 내에 있는 대상물에 촬영거리의 평균을 취해서 구한다. The figure above shows the principle of predicting the conjugate point position on the epipolar line. here

Represents the expected shooting distance of the object and is usually obtained by taking the average of the shooting distances on the object within the target area.

값으로 예상 촬영거리에 대한 최대치와 최소치를 정의한다. 그러나 실제로

값을 예측하는 것은 대단히 어렵다. 일반적인 사진에 대해서 다음과 같은 식을 이용하여 공액점의 근사 위치를 추정할 수 있다.Meanwhile,

The value defines the maximum and minimum values for the expected shooting distance. But actually

Predicting the value is very difficult. For general photographs, the approximate location of the conjugate point can be estimated using the equation

(1.20)

(1.20)

(1.21)

(1.21)

(1.22)

(1.22)

는 촬영기선길이,

는 주점기선길이,

는 점

의 시차를 나타낸다.here,

The length of the camera,

Is the tavern line length,

Point

Indicates the time difference.

를 결정한다. 연직사진이라고 가정하면 시차방정식은 다음과 같다.Search Area Using Parallax Equation

Determine. Assuming a vertical picture, the parallax equation is as follows.

(1.23)

(1.23)

: 점L의 시차here,

: Time difference of point L

: 점U의 시차

: Time difference of point U

: 점U와 L의 시차차

: Time difference between point U and L

는 다음 식으로 표현된다.So search area

Is expressed by the following equation.

(1.24)

(1.24)

(1.25)

(1.25)

(1.26)

(1.26)

(1.27)

(1.27)

는 주점기선길이(

),

는 기준면에서 촬영거리,

는 점

에서의 촬영거리,

는 대상물의 예상 촬영거리,

는

의 추정범위를 나타낸다.here

Is the tavern line length (

),

Is the shooting distance from the reference plane,

Point

Distance from,

Is the estimated shooting distance of the object,

Is

It shows the estimated range of.

   The maximum and minimum distances on the stereoscopic model can be set by converting the maximum and minimum distances assumed for the actual object as the ratio of the focal length and the photographing altitude.

   Target space image registration can be performed by setting the maximum and minimum distances of the stereoscopic model from the right image to the search area based on the initial value of the model coordinates of each feature point, and performing matching by correlation coefficient image registration. have.

The stability and accuracy of the system are improved by separating the internal and external expression processes using the extended collinear condition from the internal and external expressions using the existing DLT equation.

6 is a view briefly expressing the improvement of the system facial expression process,

The internal expression accurately calculates the principal movement of the camera, the focal length, and the lens distortion coefficient. This internal expression element can be used continuously once it is determined by the camera unique value. The internal expression work was calculated using the existing indoor model board and the calculated internal expression elements are shown in the table below.

left side  xp = 0.056708 mm yp = -0.074506 mm f = 14.357671 mm xk1 = 0.00031947136450 xk2 = 0.00000180604403 xk3 = -0.00000001609305 yk1 = 0.00033444832411 yk2 = 0.00000126198297 yk3 = -0.00000001255092 right  xp = 0.027394 mm yp = 0.000894 mm f = 14.418808 mm xk1 = 0.00054136263417 xk2 = -0.00000145609288 xk3 = -0.00000000237742 yk1 = 0.00057959190347 yk2 = -0.00000245105580 yk3 = 0.00000000356984

   The external expression, on the other hand, accurately calculates the position and attitude of the two cameras in the stereo camera system. Since these external expression factors are unique for each system, they can be used continuously once determined, but must be re-expressed if they are distorted by external environmental factors. The external expression work calculates facial expressions by installing a target outdoors and the calculated external expression elements are shown in the table below.

left side  Xo = 2860.98 mm Yo = -76.75 mm Zo = 6417.96 mm ω = -0.016091 rad φ = 0.085359 rad κ = 0.023859 rad right  Xo = 3758.04 mm Yo = -59.11 mm Zo = 6333.517 mm ω = -0.027395 rad φ = 0.094415 rad κ = 0.021807 rad

On the other hand, Figure 1 is an exploded perspective view and photograph of the damage information calculation system according to the present invention,

Damage information calculation system according to the present invention,

A crossbar 4 formed in the transverse direction, two cameras 1 arranged to be spaced apart from each other by a predetermined distance from the top of the crossbar 4 and facing the front, and the crossbar 4 from the ground A computer (2) configured to receive and store a support (3) for supporting and image information photographed by the camera (1), and calculate a range (distance and area) and damage compensation amount of the damage area by a program stored therein. It consists of.

On the other hand, the camera 1 is installed to be spaced apart by a predetermined interval on the top of the crosspiece 4, is coupled by a coupling means 5, such as a quick shoe to facilitate separation and coupling.

The computer 2 preferably uses a notebook type in consideration of mobility, and in order to receive image information photographed by the camera 1, the computer 2 must be connected by wire from the camera 1.

In order to transfer the image information captured by the camera 1 to the computer 2, the USB connection cable is not too long, and to compact the equipment, the computer 2 is connected between two cameras 1 as shown in FIG. It is preferable to configure so as to be located at.

That is, the computer 2 is installed on the worktable 6 located at the center of the crosspiece 4, and a separate USB cable (not shown) is connected between the computer 2 and the camera 1, respectively. Lose.

In addition, the worktable 6 is made of a material of iron, by combining the magnet (7) to the place where the worktable (6) is located on the crosspiece (4), the installation and dismantling work of the equipment is made easy It can be configured to

The support 3 is preferably of a type that can be folded and retracted, rather than a fixed type in consideration of mobility, it may be used in the form of a tripod, such as a camera tripod, as in the illustrated example.

On the other hand, in order to facilitate the separation and coupling between the support (3) and the crosspiece (4), the coupling means 8, such as a quick shoe is coupled to the upper end of the support (3).

Since the crossbar 4 may cause sagging due to the weight of the camera 1, it is preferable to use a rigid plate made of aluminum.

On the other hand, since the damage information extraction of the damage structure using the image capturing system is extracted from the image data, the image capturing method is important to increase the accuracy and convenience of the image processing process. The imaging method is largely dependent on the base length between the cameras, the setting of the shooting angle, the position and the posture of the shooting system. In each case, there are advantages and disadvantages, and it is necessary to shoot according to some essential shooting guidelines.

In the case of digital cameras, there are various options for image capturing. Among them, the image resolution and focus are directly related to the image expression.

   The distance between the two cameras (camera base length) is usually fixed at about 1m. From a geometric point of view, the baseline length is most accurate when the rays from two cameras are orthogonal at the location of the target point to be acquired.

Therefore, when the shooting distance is generally about 10m to 50m, the baseline length should be as large as this, which is the best in terms of accuracy. However, due to the physical limitations of the shooting system, the baseline length of more than 1m is impossible because it is very inconvenient to move and shoot, so this system is set to about 1m.

In addition, if the length of the base line is very long, the surface of the object seen in the left and right images may be different from each other, and thus the probability of image registration failure due to the correlation coefficient becomes very high in image processing using automatic image matching, and the accuracy of matching is difficult to expect.

Therefore, if the shooting distance is very distant (more than 100m), it is desirable to increase the baseline length to the maximum in order to increase the position accuracy, and to make the standard (1m) in the normal case, and to minimize the case of very close shooting (within 10m). It is impossible to freely adjust the length of the baseline because the facial expressions change every time, and only three steps can be limited to provide two facial expression files.

The shooting method according to the camera shooting angle is divided into three types. Vertical horizontal shooting is the most common shooting method with the shooting axis of two digital cameras pointing at right angles to the shooting baseline. This is the recommended method for this shooting system.

The horizontal angle shooting is a method in which two camera shooting axes are deflected left or right by a predetermined angle with respect to the base line. In this case, a device capable of accurately reading the angle of the camera should be installed.

Converged horizontal photography is a method in which two camera photographing axes are photographed at an angle converged inwardly by a predetermined angle with respect to a photographing base line, and the optical axes of the cameras cross each other.

Such a photographing method obtains the same effect as that of the photographic base line, so that very high accuracy is obtained. However, there are disadvantages of determining the convergence angle in consideration of two image redundancies and increasing the probability of failure of image registration because the area of overlapping stereoscopic images varies depending on the shooting distance.

Occation Shooting angle setting Standard Vertical horizontal If there are obstacles Declination When high accuracy of proximity object is required Convergence

   The posture of the imaging system should be level on the ground using a level meter. It is necessary to keep the ground level so that the information on the elevation difference of the object can be obtained accurately, and the projected area can be calculated. If it is not leveled, the extracted three-dimensional coordinates of the target point are calculated as the coordinates of the ground coordinate system inclined as much as the photographing system is inclined.

In this case, if the inclination angle is not measured, the calculation of the relative distance between the target points is not a problem, but the exact elevation difference and orthogonal area cannot be calculated.

   When shooting the site, the damage target exists very long, but due to the problems of the site's features (mountains, rivers, buildings), the shooting distance is not secured, so it may occur that only a part of the target cannot be taken. In this case, part of the object should be photographed while moving in succession about 30%.

 Meanwhile, the image information photographed as described above is transmitted to the computer 2, and the damage range (distance and area) and the damage amount are easily calculated by the damage range and damage calculation program stored therein.

In the damage estimation program used in the present invention, the damage amount is calculated by acquiring the damage image data from the photographing system of the damage site and input the same into the damage calculation program to finally calculate the damage estimate.

In other words, the damage volume was calculated from the input damage image data, the damage cost was calculated by applying the damage unit cost, and the damage situation and damage amount were systematically stored by storing them in the disaster data D / B.

   This is a significant improvement of the existing damage calculation method. At present, it is required to directly measure the length and area of the damage at the site to obtain the damage, and to calculate the damage by referring to the damage unit price table and fill it in the disaster register.

Therefore, it takes a lot of damage investigation time when surveying the length and area at the site, and it is difficult to arrange the damage amount and damage situation by region and type.

However, the damage calculation program proposed in the present invention may considerably reduce the time required for damage investigation by programming such a series of damage calculation, damage calculation and data cleanup.

 3 is a monitor screen picture of the damage scale calculation program according to the present invention, Figure 4 is a flow chart of the damage scale calculation program operating procedure according to the present invention.

First, in order to calculate the damage estimation amount, buttons according to a series of processes were sequentially performed as shown in item ① as shown in FIG. 4. In order to proceed, first input the damage situation, upload the damage image acquired at the damage site, and activate the damage quantity program to calculate the damage quantity.

Based on the estimated damages, the damage estimate is calculated and entered into the database. The input database can be printed according to the summary and output buttons.

   In item ②, the damage status indicates data on the damage that should be input by the author at least at the damage site, and the items include the author, river name, management office, damage date and time, damage area, and cause of damage. . In particular, an input such as damage date and time is displayed to the user only when necessary by displaying the electronic calendar as a hiding ability.

   Item ③ is the part that contains the damage image, and it will have the effect of enhancing the reality by inputting the image information collected at the damage site. In addition, the user can use the damage estimation program more conveniently because the user can view the zoom in, zoom out and location information of the damage image.

   In case of item ④, the contents of the embankment type are displayed and the dike grade, dike material, and damage length are entered.

   Item ⑤ is the part that calculates damage estimation amount based on the input damage quantity. In this part, the damage cost is built into the program based on the "National Disaster Investigation and Recovery Plan Establishment Guide" as shown in the following chart, which has the advantage of being easy to calculate the damage amount.

According to the existing method, an internal work using a separate Excel and a calculator is required, but using this program has an advantage of estimating the damage estimation amount immediately at the damage site. Therefore, it is considered to be a very improved damage estimation method in terms of accuracy and speed.

* River bank damage unit price list (Unit: KRW)

Ball bell standard unit unit price Remarks River Dike Construction -National Rivers (Festivals + Blocks)-(Festivals + Blocks + Gabions)-(Festivals + Blocks + Buddhism)-Local Class 1 Rivers (Festivals + Blocks)-(Festivals + Blocks + Gabions)-(Festivals + Blocks + Buddhism)-Local Class 2 River (Festivals + Blocks)-(Festivals + Blocks + Gabions)-(Festivals + Blocks + Saddles) m 〃 〃 〃 〃 〃 〃 〃 〃 1,412,110 1,532,950 1,810,030 1,097,650 1,266,720 1,604,990 633,740 762,040 975,450

Item ⑥ represents a database for systematically managing the contents of damages entered and calculated above. Since this database is used on a personal PC basis, the access program can be used relatively simply. Therefore, in order to use this program, MS Office 2003 version should be built in PC.

   Item ⑦ shows the function that can show the basis for estimating the damage estimate. At this time, if you press the OK button, you can see the damage estimate calculation basis. If you press the Cancel button, the screen is converted back to the damage picture.

Item ⑧ corresponds to the data processing part in the database. Its main function is to insert, delete, save, and sort data. As described above, this database is composed of access and can be modified directly in the input window and D / B. In particular, in the classification categories of data, a function to classify the data by damage type, author, damage date and time, and damage size was added in consideration of the user's convenience, so that many data can be processed in a batch.

   In the future, the statistical function will be added as a way to improve and develop the classification function. By using these statistical functions, the damage status by river, author, damage date and time and damage scale is displayed in a chart or a figure, which makes it easier to understand users.

   Item ⑨ is designed to consider the user's convenience by moving to another model of the program or performing a closing function.

  The damage amount calculation program configured as described above makes it possible to easily estimate the damage range and damage amount based on the image information photographed by the chimera.

1 is an exploded perspective view and photograph of a damage information calculation system according to the present invention;

Figure 2 is a photograph of the indoor model plate for the internal expression in the present invention,

3 is a monitor screen picture of the damage scale precision calculation program according to the present invention,

4 is a flowchart of a damage scale precision calculation program operating procedure according to the present invention;

5 is a view schematically illustrating a damage calculation process according to the present invention.

* Explanation of symbols for the main parts of the drawings

1: camera 2: computer

3: support 4: crossbar

5: coupling means 6: work table

7: magnet 8: coupling means

9: horizontal line 10: vertical line

11 rod 12 target point

14: face plate

Claims (14)

In the method of measuring the distance and area by a digital camera, Photographing objects with two cameras having the same image resolution and focal length, but partially overlapping each other; A process of determining the internal expression factors (the amount of movement of the principal point of the camera, the focal length, and the lens distortion coefficient) according to the characteristics of the camera based on the captured image data; External expression elements of the camera (

) And internal distortion factors including distortion correction coefficients are determined using the extended collinear condition equation to calculate the object coordinates in the object space from the point on the photo, and the same point in the left and right pictures by two cameras.

and

If we find, we determine the three-dimensional position by forming a linear equation with the external expression element as a variable, Comprising a step of calculating the distance or area between the two points from the determined three-dimensional position method for measuring the distance and area by a digital camera. The method of claim 1, The effects of image distortion among the camera's internal components are caused by radiation distortion, eccentric distortion, and atmospheric refraction, and are mathematically modeled as distortion correction coefficients. The coefficient to correct the distortion in the radial direction is expressed as

(From here,

Is the coefficient for radial distortion and r is the radial distance from the main point) The eccentric distortion component in the x and y directions is expressed as

Where P1 and P2 are coefficients of eccentric distortion for the combined lens at infinite focal length. The distortion correction coefficient combining radiation and eccentric distortion is

as,

(here,

: Image coordinate,

: Location of pub,

: Focal length,

: Rotation matrix,

: Object coordinates,

: Position of lens projection center) A method for measuring distance and area by a digital camera, comprising calculating an object coordinate in an object space from a point on a photograph by the above collinear conditional equation. The method of claim 1, Method for measuring the distance and area by a digital camera, characterized in that the video is taken to be partially overlapped by the two cameras 80 to 90% of the degree of overlap. In the method for calculating the range and damage amount of the damaged area by using the distance and area measurement method according to any one of claims 1 to 3, The distance or area of the damaged area is calculated from the image data photographing the damaged area by two cameras, The calculated distance or area is calculated by multiplying the cost of damage to the location, the method of calculating the range and the amount of damage area by a digital camera. The method of claim 4, wherein The calculation of the damage range (distance and area) and the damage amount is calculated by the damage calculation program stored in the computer. In the system for calculating damage information based on the image information taken with a digital camera, A cross-section formed to extend in the lateral direction, two cameras spaced apart by a predetermined distance from the top of the cross-section and facing each other, a support for supporting the cross-section from the ground, and an image photographed by the camera And a computer configured to receive and store information, and calculate a range (distance and area) of damage area and a compensation amount by a program stored therein. The method of claim 6, The camera is installed so as to be spaced apart by a predetermined interval on the top of the cross, system for calculating the damage information by the image information taken with a digital camera, characterized in that coupled by a coupling means such as a quick shoe to facilitate separation and coupling . The method of claim 6, The computer is a notebook type, the system for calculating damage information by the image information taken with a digital camera, characterized in that connected to the wired or wireless from the camera to receive the image information taken by the camera. The method of claim 8, In the case of a wired connection method, the computer is located between two cameras, and the computer is installed on a workbench located at the center of the crossbar, and a separate USB cable is connected between the computer and the camera. A system for calculating damage information based on image information taken with a digital camera. The method of claim 9, The workbench is made of iron material, and the crossbar is coupled to the magnet to the site where the workbench is located, the damage caused by the image information captured by the digital camera, characterized in that the equipment and dismantling work of the equipment is configured to be easily performed Information calculating system. The method of claim 6, The distance between the two cameras (baseline length) is increased as much as possible to increase the position accuracy when the shooting distance is very long (more than 100m), and in the normal case, the standard (about 1m), and when shooting very close (within 10m) ) Is to minimize the damage information by the digital camera, characterized in that configured to be adjusted in three steps. The method of claim 6, In the computer, through the monitor screen, the function for inputting the damage situation and activating the damage quantity program, A function for estimating damage estimation amount based on the estimated damage quantity and inputting it to the database; Data on the damage status, such as the author, river name, management agency, damage date and time, damage area, cause of damage, etc., A function of estimating the damage estimation amount based on the damage amount inputted, Database function for systematically managing the data entered in this way, It is possible to insert, delete, save, and classify data, and to classify the data by damage type, author, damage date and time, damage size. Damage information calculation system. In the digital image facial expression plate used in the distance and area measuring method according to claim 1, On the face of the flat face plate, a plurality of horizontal lines and vertical lines are displayed at equal intervals, and a plurality of rods having different lengths are installed at the intersection points of the horizontal lines and the vertical lines, and the target points are displayed at the ends of the rods. Digital facial expression board. The method of claim 13, The face plate is installed to face the front, Separately composed of two plates for portability, but connected by a hinge in the middle, the digital image facial expression plate, characterized in that the rod is coupled to the expression plate to be easily removable.

Images