• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.451-452
• /
• 2022

본 논문에서는 딥러닝 영상인식 기술을 활용한 객체검출 모델인 YOLOv4를 활용하여 탈모 자가 진단 앱을 제안한다. 본 논문에서 제안하는 앱은 실시간 영상처리기술인 YOLOv4를 사용하여 탈모 유무와 탈모 유형에 대해 학습을 하고, 앱에서 사용자가 자신의 이마 라인을 촬영하여 사진이 서버에 전송이 되고 서버에서 학습된 모델을 이용하여 검출된 탈모 유무 판단과 탈모 단계 판단의 결과값을 다시 앱으로 전송한다. 탈모에 대한 정확한 진단을 통하여 치료 방법과 치료 시기 결정에 대하여 도움을 줄 것으로 기대된다.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 1999.12a
• /
• pp.52-58
• /
• 1999

본 연구에서는 국내의 항공사진측량에서 적용되지 않고 있는 광속조정 Program을 구현하였다. 또한, 구현된 프로그램을 이용하여 과천지역에 대한 측량자료를 처리하였다. GPS를 이용하여 결정한 카메라 노출중심 위치를 이용하여 광속조정에 적용한 결과는 지상기준점수가 감소하여도 정확도는 완만하게 저하되었으나, 지상기준점만을 이용하여 광속조정에 적용한 결과는 지상기준점수가 감소함에 따라 정확도는 급격하게 저하되었다. GPS를 이용하여 결정한 카메라 노출중심 위치를 광속조정에 적용한 결과 블록 가장자리에 배치한 4개의 지상기준점 만으로 공공측량에 적합한 결과를 얻을 수 있었다. 본 연구를 통하여 GPS로 결정한 카메라의 노출중심위치를 항공삼각측량에 적용할 경우 필요한 지상기준점수를 축소함으로써, 비용을 절감할 수 있을 것으로 판단된다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.1453-1456
• /
• 2011

타일로 분할된 항공 라이다 데이터로 부터 데이터마이닝 기법을 이용한 지표면 분류 결과에 따라 건물을 포함하는 타일에 대해 적용할 건물영역 추출 기법을 제안한다. 본 기법은 재귀적 경계점 추출 알고리즘과 경계점 연결을 통해 경계선을 형성하고 경계선을 타일의 외벽과 연결해 건물영역의 외곽을 추출한다. 제안된 기법으로 추출된 건물 영역을 실제 항공사진과 비교하여 제시하고 재귀적 경계점 추출 알고리즘의 실행시간을 단축시키기 위해 사용된 지형정보 인덱스의 실행시간 단축 효과 분석이 제시된다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.23 no.3
• /
• pp.323-330
• /
• 2005

More than ten years after the era of GPS-Photogrammetry which could provide us only three projection center of all six exterior orientation parameters, direct georeferencing with GPS/INS is now becoming a standard method for image orientation. Its main advantage is to skip or reduce the indirect ground control process. This paper describes the experimental test results of integrated sensor orientation with a commercial GPS/IMU system to approve its performance in determination of exterior orientation. For this purpose two different imaging blocks were planned and the area was photographed at a large photo scale of 1:5,000 and a medium photo scale of 1:20,000. From these data set a variety of meaningful results was acquired, i.e., the accuracy. potential of exterior orientation from direct georeferencing and combined block adjustment using these data considering different photo scales and conditions.

• Journal of Korean Society for Geospatial Information Science
• /
• v.22 no.4
• /
• pp.143-149
• /
• 2014

The Disaster planning for the reservoir should be more quickly and intuitively establish measures by means of the sequential monitoring of change status of the reservoir water level and water surface area. This paper presents an approach using the orthophoto image produced by the periodic unmanned aerial photogrammetry and analyzed the feasibility. Total three time of unmanned aerial survey were conducted to make orthophoto images for the Seongnae reservoir and we analyzed the amount of changes for water level and surface area compare with each images. As the Analysis results, it was possible to effectively observe the increase in the water level rises and the surface area due to the rainfall. The maximum deviations of orthophoto images was 7.5cm in X-direction, 10.8cm in Y-direction and 14.1cm in elevation compare with ground surveying results. Therefore, we conclude that the unmanned aerial photogrammetry could be applied with comprehensive reservoir monitoring works for disaster management for reservoir in the future. And, the orthophoto production takes about two hours to shoot the images, and approximately four hours is considered for the image processing. So, the unmanned aerial photogrammetry is considered to be the best disaster work that requires urgent because analysis is possible in the shooting day.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.19 no.4
• /
• pp.357-363
• /
• 2001

Recently, structure and type of GIS data is various and complicated. Therefore, there is a research on efficient data collecting and processing method. Digital maps and orthophotos are needed for efficient usage in the field of GIS. Digital maps which are one of the part of NGIS and digital cadastral maps are different, and data construction isn't performed. Furthermore, there are no research on the improvement of difference between digital maps and cadastral maps. In this paper, manufactured digital orthophotos using aerial photos are used for change detection in a central area of a town. Selecting a town which has various changes, and digital maps and cadastral maps overlap each other, and analyze them for correction, renovation and possible application.

• Journal of Korean Society of Forest Science
• /
• v.96 no.3
• /
• pp.251-258
• /
• 2007

Airborne Lidar (light detection and ranging) can be an effective alternative in forest inventory to overcome the limitations of conventional field survey and aerial photo interpretation. In this study, we attempt to develop methodologies to identify individual trees and to estimate tree height from airborne Lidar data. Initially, digital elevation model (DEM) data representing the exact ground surface were generated by removing non-ground returns from the multiple-return laser point clouds, obtained over the coniferous forest site of rugged terrain. Based on the canopy height model (CHM) data representing non-ground layer, individual tree heights are extracted through pseudo-grid method and moving window filtering algorithm. Comparing with field survey data and aerial photo interpretation on sample plots, the number of trees extracted from Lidar data show over 90% accuracy and tree heights were underestimated within 1.1m in average at two plantation stands of pine (Pinus koraiensis) and larch (Larix leptolepis).

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.9 no.3
• /
• pp.155-164
• /
• 1997

A comprehensive and systematic field monitoring program was initiated since October 1989, in order to investigate the temporal and spatial variation of shoreline position at northern part of Pea Island, North Carolina. Aerial photographs were taken every two months on the shoreline extending from the US Coast Guard Station at the northern end of Pea Island to a point 6 miles to the south. Aerial photographs taken were digitized initially to obtain the shoreline position data. in which a wet-dry line visible on the beach was used to identify the position of shoreline. Since the wet-dry line does not represent the “true" shoreline .position but includes the errors due to the variations of wave run-up heights and tidal elevations at the time the photos taken, it is required to eliminate the tide and wave runup effects from the initially digitized shoreline .position data. Runup heights on the beach and tidal elevations at the time the aerial photographs taken were estimated using tide data collected at the end of the FRF pier and wave data measured from wave-rider gage installed at 4 km offshore, respectively A runup formula by Hunt (1957) was used to compute the run-up heights on the beach from the given deepwater wave conditions. With shoreline position data corrected for .wave runup and tide, both spatial and temporal variations of the shoreline positions for the monitoring shoreline were analyzed by examining local differences in shoreline movement and their time dependent variability. Six years data of one-mile-average shoreline indicated that there was an apparent seasonal variation of shoreline, that is, progradation of shoreline at summer (August) and recession at winter (February) at Pea Island. which was unclear with the uncorrected shoreline position data. Determination of shoreline position from aerial photograph, without regard to the effects of wave runup and tide, can lead to mis-interpretation for the temporal and spatial variation of shoreline changes.nges.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.11a
• /
• pp.109-112
• /
• 2006

이미지 모자이크 기법은 한 번에 촬영 할 수 없는 큰 배경이나 사물을 부분적으로 촬영 한 후 이들을 조합하여 전체 배경이나 사물을 합성하는 기술이다. 이 기술은 주로 지형을 촬영한 항공사진을 조합하여 전체 영상을 얻는 용도로 사용되어왔다. 본 연구에서는 일반적인 스틸사진의 조합이 아닌 동영상으로 촬영된 영상물로부터 전체 배경을 합성하는 방법에 대한 것이다. 이를 위하여 먼저 비디오 프레임간의 공통적인 특징 점들을 추출하고 일치되는 점들을 찾아내었다. 이로부터 두 프레임간의 상대적인 좌표를 구한 후 이를 저장하였다. 마지막으로 합성 단계에서는 저장된 상대 좌표로부터 부분 이미지를 연결하여 전체 이미지를 구하는 방법을 제안하였다. 제안된 방법은 근접 촬영이 필수인 협소공간의 고대 구조물의 육안 검사를 소형로봇 등의 원격 이동체에 실린 카메라로 수행하기 위한 용도에 응용될 수 있다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.6 no.2
• /
• pp.34-41
• /
• 1988

Nowadays the considerable improvements, not only in photogrammetric equipment but also in adjustment methods have led to a high accuracy of Aerial Triangulation and reach the step to obtain A.T. technique that uses Global Position System without ground control survery. The enterance of analytical plotter have made considerable development both in accuracy and in efficiency and improved working environment of A.T. by the combination of On-Line Analytical Plotting System. In this paper, in A.T., with using of analytical polotter, aims to improve the acccuracy in coordinates determination of low level control point. The result shows that in case of A.T. with analytical plotter AVIOLYT BC2, the accuracy of control points improves to 7% in planimetry and to 40% in height compared to A.T. with precision plotter Autograph A-8.