• The Korean Journal of Ecology
• /
• 제24권6호
• /
• pp.385-394
• /
• 2001

An attempt to clarify the landscape structure of urban areas was carried out in the greenbelt around Seoul, Korea's metropolis. By means of aerial photographs and a field survey, a vegetation map including land-use pattern was made. Landscape structure was described by analyzing this vegetation map and the results of phytosociological survey. Landscape element types identified were (1) secondary forest, (2) plantation, (3) cultivated field, (4) urbanized area, (5) graveyard, and (6) bare rock. Vegetation units, resulting from the phytosociological analysis, included Quercus mongolica, Q. variabilis, Q. acutissima, Pinus densiflora, Q. aliena, and Alnus japonica communities. Plantations were composed of Robinia pseudoacacia, Populus tomentiglandulosa, P. rigida, Larix leptolepis, P. koraiensis, and Castanea crenata stands. Patches near to human settlements in the lower zones of the mountains were fragmented and small but they became larger towards the higher mountain zones. On the other hand, the number of patches was fewer and their size was larger in Mt. Cheonggye more distant from the principal residential area, larger in size, and higher in elevation compared with the other 2 mountains, Mt. Daemo and Mt. Acha. Floristic composition of Mongolian oak(Q. mongolica) stand distributing in the upper part of each mountain, in which artificial interference is rare, showed a difference among those study areas different in parent rock and disturbance regime. But that of black locust(R. pseudoacacia) stand located in lowland of mountainous area, in which artificial interference is frequent was similar to each other. As the results of analyses on the frequency distribution of diameter classes of major species, dominant landscape elements, Mongolian oak forest showed different responses depending on artificial interference as continuous maintenance and retrogressive succession in the sites far from and near to the residential areas, respectively. On the other hand, black locust stands showed a probability to be restore to the native oak forest through progressive succession.

• 지질공학
• /
• 제26권1호
• /
• pp.23-32
• /
• 2016

ASTER 위성영상을 이용하여 팽창성 점토광물인 일라이트 인자 추출 및 SVM 통계분석을 통해 산사태 취약성을 평가하였다. 연구지역의 산사태 발생지역은 항공사진 판독 및 현장 조사를 통해 분석하였다. GIS 기반 공간데이터베이스로는 지형도, 토양도, 임상도, ASTER 위성사진을 이용하였다. 수치지형도에서는 경사 및 경사방향, 곡률도, 계곡과의 거리, 도로와의 거리, 토양도에서는 유효토심, 토질, 토양지형, 토양 배수정도 및 토양 모재, 임상도에서는 경급, 영급 및 밀도를 위성사진에서는 일라이트 인자를 추출하였다. 산사태 발생요인 데이터베이스와 SVM 통계분석 및 가중치 계산을 통해 각 요소간의 상관관계 취약성도를 구하였다. AUC 검증 결과 일라이트 인자 적용결과는 76.46%의 예측 정확도를 보였으며 일라이트 인자 미적용 모델은 74.09%의 예측 정확도를 나타내었다. 이는 일라이트 인자가 산사태 취약성도 작성에 있어 중요한 자료로 사용될 수 있음을 나타낸다.

• 대한원격탐사학회지
• /
• 제33권4호
• /
• pp.423-436
• /
• 2017

This paper presents a deep learning-based road segmentation framework from very high-resolution orthophotos. The proposed method uses Deep Convolutional Autoencoders for end-to-end mapping of orthophotos to road segmentations. In addition, a set of post-processing steps were applied to make the model outputs GIS-ready data that could be useful for various applications. The optimization of the model's parameters is explained which was conducted via grid search method. The model was trained and implemented in Keras, a high-level deep learning framework run on top of Tensorflow. The results show that the proposed model with the best-obtained hyperparameters could segment road objects from orthophotos at an average accuracy of 88.5%. The results of optimization revealed that the best optimization algorithm and activation function for the studied task are Stochastic Gradient Descent (SGD) and Exponential Linear Unit (ELU), respectively. In addition, the best numbers of convolutional filters were found to be 8 for the first and second layers and 128 for the third and fourth layers of the proposed network architecture. Moreover, the analysis on the time complexity of the model showed that the model could be trained in 4 hours and 50 minutes on 1024 high-resolution images of size $106{\times}106pixels$, and segment road objects from similar size and resolution images in around 14 minutes. The results show that the deep learning models such as Convolutional Autoencoders could be a best alternative to traditional machine learning models for road segmentation from aerial photographs.

• 대한토목학회논문집
• /
• 제12권1호
• /
• pp.151-158
• /
• 1992

항공사진보다 경제적인 SPOT 위성의 입체영상으로부터 지형도를 제작하는 것이 가능하게 되었다. 많은 학자들이 국가기본도 수정에의 적용가능성과 결과의 정확도를 분석하여 SPOT 영상은 1/50,000에서 1/100,000까지의 지형도에 적합하다고 발표하였다. SPOT 영상으로부터 지형도, 정사사진 및 수치지형모델을 제작하기 위해서는 정확한 외부표정요소를 알아야 하지만, 동적인 선형주사 영상인 SPOT 영상의 특성 때문에 사진측량에서 이용되어온 종래의 번들조정은 직접 사용되지 못한다. 신뢰성은 내적신뢰성으로 불리는 과대오차의 검출능력과 외적신뢰성으로 불리는 검출되지 않은 과대오차가 외부표정 결과에 미치는 영향을 나타낸다. 본 연구에서는 좌표계, 궤도방향 좌표의 표현방법, 외부표정 요소의 차수 및 기준점 배치에 따라 SPOT 영상의 신뢰성이 어떻게 달라지는가를 밝히고, SPOT 영상을 이용한 작업에서의 기초자료로 제공될 수 있는 외부표정에서의 이론적인 신뢰성을 분석하였다.

• 한국환경복원기술학회지
• /
• 제4권2호
• /
• pp.36-44
• /
• 2001

The main purpose of this study was to classify forest vegetation effectively using Landsat Thematic Mapper data(June, 1994) in mountainous region. The research area was the Mt. Wolak Experimental Forest of Chungbuk National University, near Chungju and Jecheon city, Chungcheongbuk-do. To classify forest vegetation effectively, Normalized Difference Vegetation Index(NDVI) was used to reduce topographic effects. This NDVI was modified and transformed to the value of 0 to 255, and then the modified values were combined with other Landsat Thematic Mapper bands. To classify forest and land cover types, unsupervised classification method was used. The results of this study are summarized as follows. 1. Combinations of band "3, 5, NDVI" in Landsat Thematic Mapper data showed a good separation with high accuracy. The expected classification accuracy was 95.1% in Landsat Thematic Mapper data. 2. The Land Cover types were classified into six groups : coniferous forest, deciduous forest, mixed forest, paddy and grass, non-forest, and other undetectable areas. As these classified results were compared with the reconnaissance survey and aerial black and white infrared photographs, the overall classification accuracy was 76.5% in Landsat Thematic Mapper data. 3. The portion of non-forest in Mt. Wolak area was 1.9%. The percentages of coniferous, deciduous and mixed forests were 30.9%, 35.7% and 26.4%, respectively. 4. As these classified results were compared with other reference data, the percentages of coniferous, deciduous and mixed forests increased, but the portion of non-forest was exceedingly diminished. These differences are thought to be from the different research method and the different season of received Landsat Thematic Mapper data.

• 한국농촌건축학회논문집
• /
• 제20권3호
• /
• pp.11-18
• /
• 2018

Suhwa Kim Hwan-gi established his own unique world of art based on Korean-style lyricism with his sophisticated and sublimated formative language as the first generation of Korean abstract artist. He made his name known not only in Korea but New York and Paris, the center of contemporary art. Presently, there is almost no material left for us to examine how Kim Hwan-gi's house looked in the first place. But there is a painting that chef Kim Am-gi possesses where we can find the image of painter Kim Hwan-gi's house drawn by recalling the memories at home in a far-off land. With that, we can see it should be the original looks of that house. To examine the original looks of Kim Hwan-gi's house when it was built, this author studied the closure land registration map. By analyzing 29 land registration maps around Eupdong-ri of Gijwa-myeon in Muan-gun of Jeollanam-do produced in 1917 which National Archives of Korea possesses, this researcher could figure out the status of land registration around Kim Hwan-gi's house with No. 15 and 22 drawing boards. According to the results of considering aerial photographs after the emancipation that National Geographic Information Institute possesses, this author has found that the arrangement of Kim Hwan-gi's house was changed from 1954 up to 2008. In 1954 aerophotos, we can see the main building and studio in Kim Hwan-gi's house and there was an additional structure between them. Kim Hwan-gi's house was built before 1925. The main building presently designated as a cultural asset and also the sarangchae arranged in the east and the studio, too, were built in the similar period.

• 한국수자원학회논문집
• /
• 제41권4호
• /
• pp.395-404
• /
• 2008

충적하천의 복잡하고 다양한 이동특성을 파악하고 이해하는 것은 하천공학적으로 매우 중요하며, 본 연구에서는 저수로의 이동 및 하상저하로 인하여 취수문제가 있는 경상북도 구미시에 위치한 해평취수장 주변에 대하여 항공사진 분석을 통하여, 시간에 따른 하천의 지형변화, 저수로의 이동 특성을 조사하고 분석하였다. 저수로는 좌안에서 우안으로 이동해 가고 있으며, 저수로는 강턱유량에 대하여 복렬사주가 발달하는 특성을 보여주고 있다. 이는 하천의 경사가 급하고 하폭이 넓고 수심이 얕은 곳에서 발생하는 현상으로서, 저수로의 불안정성이 증가하기 때문이다. 또한 저수로의 사행도가 감소하면서 저수로 폭이 증가하는 것은 제한된 범위 안에서 저수로의 분류와 합류가 활발하게 발생하는 것을 의미한다. 시간의 증가에 따라 저수로 하폭의 증가율과 측방향 이동율은 감소하고 있다. 강턱유량을 이용하여 중규모 영역구분을 수행한 결과, 복렬사주가 발달하는 하천의 특성을 보여주었으며, 모래하천으로서 부유사가 지배적인 하천으로 판단되었다.

• 한국환경과학회지
• /
• 제16권5호
• /
• pp.603-610
• /
• 2007

Recent urban sprawl has destroyed various kinds of green space in tile city. It has affected duality of people's life in the city, as well as urban ecosystem. Recent study shows the possibilities of roofs as green spaces in urban central site where the land costs are generally high. This research focuses on Jung-Gu district in Daegu Metropolitan city as a study area and calculates possible area of green roof using 2002 Autocad program based on aerial photographs and land registration maps. And the purpose of this research is to analyze environmental and economic effects of green roof. The environmental effects are as follows. It is expected that $91,106m^2$ green spaces, 12.13 % of study site, will be added if green roof is performed in the study site. It is assumed that the expanded areas could reduce the highest temperature to $0.5-1.0^{\circ}C$ during the summer in terms of environmental effect. And the following shows the economic effects. If green roof and greening urban central site are created as a same size of $91,106m^2$, it will be expected that the costs of green roof will be much more in-expensive than about 98 billions won. It will be also found that the expense of cooling energy can be saved out about 8 millions won per day in summer, if grass planting is accomplished on the possible areas of green roof in the study site. Therefore, it is desirable to take legal supports such as enacting regulations to activate green roof for more environmental and economic effects. For instance, green roof for public institutions, school and model area selection are desirable method to publicize the effect of greening program for citizen's participation.

• 대한토목학회논문집
• /
• 제9권4호
• /
• pp.113-124
• /
• 1989

인공위성 데이타 정보의 용용분야는 여러가지가 있으나, 본 연구에서는 LANDSAT MSS데이타와 TM데이타를 처리 분석하여 서울시 토지이용정보를 경년변화에 따라 추출하고자 하였다. 사용 데이터는 MSS(72, 79, 81, 83년), TM(85년)이며 입수된 데이타를 전처리를 통해 기하보정, 디지타이징(행정구역에 따라) 등을 하고, 유효 band 선정 및 filtering을 통하여 정확도를 높인 후 MLC(Maximum Likelihood Classifier)로 토지이용분류를 실시하였다. 토지이용분류시 training field 선정 자료로는 현지조사자료, 지형도, 항공사진을 참조하였고, 분류결과의 정확도는 각각 그 당시의 통계자료를 토대로 하여 비교해 보았다. 분석결과, 서울시의 도시지역은 72년 (25.3 %), 81년 (43.0 %), 85년 (51.9 %)로 증가되었고, 이에 대해 삼림은 72년(39.0 %)에서 85년(28.4 %)로 점차 감소되고 있었다. 이상과 같이 토지이용 경년변화를 추출함으로써 도시의 토지 이용상황 monitoring에는 반복 주기를 가지는 인공위성 데이터의 활용이 경제적이며 효과적임을 알 수 있었다.

• Spatial Information Research
• /
• 제16권3호
• /
• pp.373-383
• /
• 2008

지표면에 대한 정보를 취득하는 기법 중 지금까지 주로 사용되어온 기법은 항공사진 및 위성영상과 같이 평면적인 정보 수집에 중점을 두고 있는 반면, 본 논문에서 다루는 LiDAR(Light Detection And Ranging)는 레이저 측량기술을 이용하여 지표면에 대한 고해상도의 비정규분포 Point 형태의 3차원 정보의 획득이 가능하다. GPS(Global Positioning System) 수신기와 INS(Inertial Navigation System)의 결합을 통해 좌표 값을 제공하게 된다. 이러한 LiDAR의 3차원 Point 정보와 좌표 값을 활용하여 보다 정밀한 3차원 모델링 수행이 가능하다. 본 연구에서는 LiDAR의 반사강도와 기하/지형 자료를 이용하여 도시지역을 대상으로 정밀한 3차원 공간정보자료를 취득하고, 그 자료를 분석하여 도시지역을 높이와 밀도를 기반으로 하여 3차원으로 분류하였다. LiDAR를 통해 획득된 원시자료로부터 지표면에서 반사되는 Point Data의 개수를 지면과 비지면 요소의 비율로 추정하여 지형과 공간적 특성을 파악하고 이에 따라 3차원 토지피복분류도를 작성하였다. 신호의 강약을 구분하는 기준은 통계적 방법(Jenk's Natural Break)을 통해 추정된 값을 사용하였으며, 지표면 반사비율에 따라 세부지역으로 구분하여 크게 고밀도 저밀도 식생지역과 비식생지역으로 구분하였다.