• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2009년도 춘계학술대회 논문집
• /
• pp.286-290
• /
• 2009

In recent years, LiDAR technology has been becoming more popular and important. Its applications are completely replacing the traditional remote sensing technique. One of these applications is creating Digital City Models in urban areas, which is essential for many others such as disaster management, cartographic mapping, simulation of new buildings, updating and keeping cadastral data. In most of these cases the building outlines is the primary feature of interest. In this paper, a method of extracting building outlines from LiDAR data will be performed.

• 대한원격탐사학회지
• /
• 제28권5호
• /
• pp.549-559
• /
• 2012

본 연구는 표고자료의 임의지점에서 표면의 형상에 대한 파라미터들을 추정하는 경우, 추정에 적용된 조정계산모델들의 전통계량이 추정결과를 채택 혹은 기각하는데 적용될 수 있는 정도를 비교분석하는 것을 목적으로 한다. 표면 파라미터 추정시 그 추정결과값을 신뢰할 수 있는지에 대한 여부는 일반적으로 표면의 조건, 그리고 적용된 조정계산모델의 유형에 따라 영향을 받을 것으로 예상되지만 이에 대한 체계적이고 구체적인 연구는 아직 미흡한 실정이다. 원시자료가 실제 관측으로부터 취득된 자료인 경우, 그 참값을 알 수 없기 때문에, 조정계산모델들의 적용적합성을 명확히 파악하기 위하여, 본 연구에서는 원시자료를 모의적으로 생성하여 연구를 수행하였다. 모의자료생성은 각 기준면위에 수평면, 경사면, 그리고 곡면을 포함시키고 일정량의 노이즈를 추가함으로써 연구를 수행하였다. 비교 분석은 모의자료의 임의지점에서 추정된 결과값들을 각 조정계산모델별 전통계량과 연계분석함으로써 수행되었다. 실험결과로부터 조정계산모델들의 다양한 표면조건에 따른 추정특성을 파악할 수 있었다.

• 대한원격탐사학회지
• /
• 제29권3호
• /
• pp.275-291
• /
• 2013

Monitoring the global Gross Primary Pproduction (GPP) is relevant to understanding the global carbon cycle and evaluating the effects of interannual climate variation on food and fiber production. GPP, the flux of carbon into ecosystems via photosynthetic assimilation, is an important variable in the global carbon cycle and a key process in land surface-atmosphere interactions. The Moderate-resolution Imaging Spectroradiometer (MODIS) is one of the primary global monitoring sensors. MODIS GPP has some of the problems that have been proven in several studies. Therefore this study was to solve the regional mismatch that occurs when using the MODIS GPP global product over Korea. To solve this problem, we estimated each of the GPP component variables separately to improve the GPP estimates. We compared our GPP estimates with validation GPP data to assess their accuracy. For all sites, the correlation was close with high significance ($R^2=0.8164$, $RMSE=0.6126g{\cdot}C{\cdot}m^{-2}{\cdot}d^{-1}$, $bias=-0.0271g{\cdot}C{\cdot}m^{-2}{\cdot}d^{-1}$). We also compared our results to those of other models. The component variables tended to be either over- or under-estimated when compared to those in other studies over the Korean peninsula, although the estimated GPP was better. The results of this study will likely improve carbon cycle modeling by capturing finer patterns with an integrated method of remote sensing.

• 대한원격탐사학회지
• /
• 제35권1호
• /
• pp.93-115
• /
• 2019

The availability of high-resolution laser scanning data and advanced machine learning algorithms has enabled an accurate potential rockfall source identification. However, the presence of other mass movements, such as landslides within the same region of interest, poses additional challenges to this task. Thus, this research presents a method based on an integration of Gaussian mixture model (GMM) and ensemble artificial neural network (bagging ANN [BANN]) for automatic detection of potential rockfall sources at Kinta Valley area, Malaysia. The GMM was utilised to determine slope angle thresholds of various geomorphological units. Different algorithms(ANN, support vector machine [SVM] and k nearest neighbour [kNN]) were individually tested with various ensemble models (bagging, voting and boosting). Grid search method was adopted to optimise the hyperparameters of the investigated base models. The proposed model achieves excellent results with success and prediction accuracies at 95% and 94%, respectively. In addition, this technique has achieved excellent accuracies (ROC = 95%) over other methods used. Moreover, the proposed model has achieved the optimal prediction accuracies (92%) on the basis of testing data, thereby indicating that the model can be generalised and replicated in different regions, and the proposed method can be applied to various landslide studies.

• 대한원격탐사학회지
• /
• 제35권1호
• /
• pp.179-193
• /
• 2019

Landslides are one of the most damaging geological hazards worldwide, threating both humans and property. Hence, there have been many efforts to prevent landslides and mitigate the damage that they cause. Among such efforts, there have been many studies on mapping landslide susceptibility. Geographic information system (GIS)-based techniques have been developed and applied widely, and are now the main tools used to map landslide susceptibility. We reviewed the status of landslide susceptibility mapping using GIS by number of papers, year, study area, number of landslides, cause, and models applied, based on 776 articles over the last 20 years (1999-2018). The number of studies published annually increased rapidly over time. The total study area spanned 65 countries, and 47.7% of study areas were in China, India, South Korea, and Iran, where more than 500 landslides, 27.3% of all landslides, have occurred. Slope (97.6% of total articles) and geology (82.7% of total articles) were most often implicated as causes, and logistic regression (26.9% of total articles) and frequency ratio (24.7% of total article) models were the most widely used models. We analyzed trends in the causes of and models used to simulate landslides. The main causes were similar each year, but machine learning models have increased in popularity over time. In the future, more study areas should be investigated to improve the generalizability and accuracy of the results. Furthermore, more causes, especially those related to topography and soil, should be considered and more machine learning models should be applied. Finally, landslide hazard and risk maps should be studied in addition to landslide susceptibility maps.

• 대한원격탐사학회지
• /
• 제33권4호
• /
• pp.339-350
• /
• 2017

In recent, cloud computing paradigm and open source as a huge trend in the Information Communication Technology (ICT) are widely applied, being closely interrelated to each other in the various applications. The integrated services by both technologies is generally regarded as one of a prospective web-based business models impacting the concerned industries. In spite of progressing those technologies, there are a few application cases in the geo-based application domains. The purpose of this study is to develop a cloud-based service system for satellite image processing based on the pure and full open source. On the OpenStack, cloud computing open source, virtual servers for system management by open source stack and image processing functionalities provided by OTB have been built or constructed. In this stage, practical image processing functions for KOMPSAT within this service system are thresholding segmentation, pan-sharpening with multi-resolution image sets, change detection with paired image sets. This is the first case in which a government-supporting space science institution provides cloud-based services for satellite image processing functionalities based on pure open source stack. It is expected that this implemented system can expand with further image processing algorithms using public and open data sets.

• 한국지리정보학회지
• /
• 제27권1호
• /
• pp.99-114
• /
• 2024

국내 해안 쓰레기 모니터링은 노동 집약적 방식과 한정적인 조사 범위로 세밀한 분포 확인이 어렵다. 따라서 해안 쓰레기 자료수집의 효율성 증대를 위해 원격탐사 기법을 이용한 연구가 이루어지고 있다. 하지만 국내 원격탐사 기반 해안 쓰레기 모니터링 방안에 대한 기준이 미흡한 실정이다. 본 연구에서는 국내 연구 결과를 기초로 메타분석 방법을 적용하여 원격탐사체 중 드론을 이용한 해안 쓰레기 모니터링 연구 19건에 대해 모니터링 방법과 결과에 대해 분석하였다. 모니터링 방법을 대상으로 데이터 수집 방법, 수집 데이터 정보, 모니터링 대상지 정보에 대해 분석하였으며, 모니터링 결과를 대상으로 모니터링 실태, 탐지대상 및 활용모델에 대해 분석하였다. 또한, 메타분석 결과를 바탕으로 드론을 이용한 해안 쓰레기 모니터링 수행 시 고려 항목과 권장 항목, 수행 기준에 대한 모니터링 기준 항목을 제시하였다. 본 연구 결과를 통해 드론을 이용한 해안 쓰레기 모니터링 운용 기준 마련에 필요한 조건 및 기준을 정의하였으며, 추후 외국 사례 분석 및 현장 적용 결과를 추가하여 국가 차원의 원격탐사체를 이용한 해안 쓰레기 모니터링 지침 마련이 가능할 것으로 보인다.

• Journal of Forest and Environmental Science
• /
• 제40권1호
• /
• pp.15-23
• /
• 2024

This research aimed to assess the possibility of detecting forest degradation using time-series satellite imagery and three different deep learning-based change detection techniques. The dataset used for the deep learning models was composed of two sets, one based on surface reflectance (SR) spectral information from satellite imagery, combined with Texture Information (GLCM; Gray-Level Co-occurrence Matrix) and terrain information. The deep learning models employed for land cover change detection included image differencing using the Unet semantic segmentation model, multi-encoder Unet model, and multi-encoder Unet++ model. The study found that there was no significant difference in accuracy between the deep learning models for forest degradation detection. Both training and validation accuracies were approx-imately 89% and 92%, respectively. Among the three deep learning models, the multi-encoder Unet model showed the most efficient analysis time and comparable accuracy. Moreover, models that incorporated both texture and gradient information in addition to spectral information were found to have a higher classification accuracy compared to models that used only spectral information. Overall, the accuracy of forest degradation extraction was outstanding, achieving 98%.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.72-76
• /
• 2004

For successful route guidance, navigation systems should provide to users more realistic and actual information such as photographs than those in either 2-dimensional and 3-dimensional models. In this paper, we propose a method for servicing photographs for route guidance in navigation systems. The method includes how to acquire photographs with the most successful view for the guidance, how to construct link information among them and navigational map data, and how to provide the images to users efficiently.

• 대한원격탐사학회지
• /
• 제22권3호
• /
• pp.221-228
• /
• 2006

In this study, an integrated 3D modeling and rendering system dealing with 3D urban landscape features such as terrain, building, road and user-defined geometric ones was designed and implemented using $OPENGL\;{|}\;ES$ (Embedded System) API for mobile devices of PDA. In this system, the authoring functions are composed of several parts handling urban landscape features: vertex-based geometry modeling, editing and manipulating 3D landscape objects, generating geometrically complex type features with attributes for 3D objects, and texture mapping of complex types using image library. It is a kind of feature-based system, linked with 3D geo-based spatial feature attributes. As for the rendering process, some functions are provided: optimizing of integrated multiple 3D landscape objects, and rendering of texture-mapped 3D landscape objects. By the active-synchronized process among desktop system, OPENGL-based 3D visualization system, and mobile system, it is possible to transfer and disseminate 3D feature models through both systems. In this mobile 3D urban processing system, the main graphical user interface and core components is implemented under EVC 4.0 MFC and tested at PDA running on windows mobile and Pocket Pc. It is expected that the mobile 3D geo-spatial information systems supporting registration, modeling, and rendering functions can be effectively utilized for real time 3D urban planning and 3D mobile mapping on the site.