무인항공기를 이용한 소규모 지역의 3차원 공간정보 구축 및 갱신 = Construction and Updating of 3D Spatial Information for Small Areas using UAV

다국어 초록 (Multilingual Abstract)

With the increase in the volume of data due to rapid social development, accurate spatial information is required. This situation is leading to the conduct of studies for the construction and updating of 3D spatial information. The 3D spatial information is constructed using existing data, including digital maps, or using survey techniques, including aerial photogrammetry and the airborne laser survey.
The 3D spatial information is constructed by ‘Si’ or ‘Gun’ units. It cannot be periodically reconstructed or updated regularly because it requires much time and cost.
The 3D spatial information must be updated to apply many changes from various construction works, such as infrastructure works, buildings and housing. These changes must be quickly applied to the 3D spatial information to ensure timeliness and reliability.
The 3D spatial information can be updated using digital maps, which are updated every two years for metropolitan areas and every four years for the rest. The update of the 3D spatial information using digital maps is disadvantageous, because the changes that occur between the updates cannot be applied. Therefore, an on-demand update system is required, instead of the periodic bulk update system with two to four year intervals.
This study proposed a method of acquiring aerial images of cities where geographic features, such as terrain and objects often change, by installing a camera in an UAV(Unmanned Aerial Vehicles); and a technique for constructing or frequently updating the 3D spatial information using the acquired aerial images. Thus, this study was conducted to reduce the cost and time for updating, minimize overlapping investments, and improve the quality and reliability of 3D spatial information.
For the study, the acquisition and update methods of 3D spatial information were analyzed, and a process for acquisition and update using UAV was developed through sequence of plan set up, aerial images acquisition, data processing and result analysis.
The study areas were chosen according to the technique that was proposed in this study, and ground control point selection and a control point survey were conducted. The camera that was to be installed in the UAV was chosen, the elements of its interior orientation were acquired via the camera calibration, and the UAV and gimbal for the aerial photogrammetry were chosen.
The 3D spatial information database was constructed by orientation, bundle adjustment, and 3D-mapping by the acquired images. New 3D spatial information and missing information were analyzed by comparing them with those from the existing digital maps. The errors and location accuracy of the field tests were analyzed using the 1/1,000 digital maps and the ground control point survey.
The accuracy of the results was analyzed by applying the unmanned aerial imaging system to the field and comparing the results with those of the control point survey. The standard deviations in the X, Y, and Z directions were 0.193 m, 0.152 m, and 0.218 m, respectively. When the buildings in the area were 3D-mapped and the results were compared with the locations of the buildings and corner points in the 1/1,000 digital map, the standard deviations in the X and Y directions were 0.192 m and 0.240 m, respectively.
As for the artificial features, 108 of the 123 buildings in the subject area were mapped using the proposed technique via a field application test. As 14 buildings that were built after August 2008 were mapped, it seems that the technique can be used for on-demand updating of geospatial information.
The cost of the UAV photogrammetric method (the control point survey, imaging, and data processing) was compared with the cost of the aerial photogrammetry (imaging, control point survey, digital mapping, field survey, and digital map manufacturing). The UAV photogrammetric method cost only about 5% of the cost of the existing aerial photogrammetry.
As above, It can be concluded that developing a system in which information changes in diverse realities can be frequently updated via the aerial photogrammetric method using an UAV could reduce the time needed for updating of digital maps and 3D spatial information, and minimize the overlapping investments that occur in the existing bulk updating system, which involves even areas without changes.

목차 (Table of Contents)

• Ⅰ. 서 론 1
• 1.1 연구의 필요성 및 목적 1
• 1.2 국내외 연구동향 3
• 1.3 연구방법 및 범위 11
• Ⅱ. 3차원 공간정보와 디지털 사진측량 14
• 2.1 3차원 공간정보 14
• 2.2 디지털 사진측량의 영상기하 22
• 2.2.1 공선조건 22
• 2.2.2 공면조건 27
• 2.2.3 공액조건 30
• 2.2.4 광속조정 31
• 2.3 영상정합 32
• 2.3.1 영역기준 정합 34
• 2.3.2 형상기준 정합 42
• 2.3.3 관계형 정합 50
• 2.4 렌즈 왜곡 51
• 2.4.1 방사왜곡 52
• 2.4.2 편심왜곡 53
• Ⅲ. 무인항공기 촬영시스템 구축 및 자료취득 56
• 3.1 대상지역 선정 56
• 3.2 무인항공기 촬영시스템 58
• 3.2.1 무인항공기 개념 58
• 3.2.2 무인항공기 분류 58
• 3.2.3 무인항공 촬영시스템 구축 61
• 3.3 카메라 검정 64
• 3.4 항공사진 촬영 71
• 3.5 기준점 측량 73
• Ⅳ. 자료처리 및 결과분석 79
• 4.1 카메라 현장검정 79
• 4.2 자료처리 80
• 4.3 3차원 공간정보 갱신 88
• 4.4 결과분석 96
• 4.4.1 정확도 분석 96
• 4.4.2 공간정보 구축 및 갱신 99
• 4.4.3 효용성 분석 105
• Ⅴ. 결 론 108
• 참고문헌 110