• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1149-1161
• /
• 2021

With a trend of the utilization of computer vision for satellite images, cloud detection using deep learning also attracts attention recently. In this study, we conducted a U-Net cloud detection modeling using SPARCS (Spatial Procedures for Automated Removal of Cloud and Shadow) Cloud Dataset with the image data augmentation and carried out 10-fold cross-validation for an objective assessment of the model. Asthe result of the blind test for 1800 datasets with 512 by 512 pixels, relatively high performance with the accuracy of 0.821, the precision of 0.847, the recall of 0.821, the F1-score of 0.831, and the IoU (Intersection over Union) of 0.723. Although 14.5% of actual cloud shadows were misclassified as land, and 19.7% of actual clouds were misidentified as land, this can be overcome by increasing the quality and quantity of label datasets. Moreover, a state-of-the-art DeepLab V3+ model and the NAS (Neural Architecture Search) optimization technique can help the cloud detection for CAS500 (Compact Advanced Satellite 500) in South Korea.

• Journal of the Korean earth science society
• /
• v.40 no.3
• /
• pp.212-226
• /
• 2019

This study derived spectral Lambertian Equivalent Reflectance (LER) over East Asia from the observations of Ozone Monitoring Instrument (OMI) onboard polar-orbit satellite Aura. The climatological (October 2004-September 2007) LER values were compared with the surface reflectance products of OMI or MODerate resolution Imaging Spectroradiometer (MODIS) in terms of the atmosphere-environment variables as follows: wavelength (UV, visible), surface properties (land, ocean), and cloud filtering. Four kinds of LER outputs in the UV and visible region (328-500 nm) were retrieved based on the averages of lowest (1, 5, and 10%) surface reflectance values as well as the minimum reflectance. The average of the lowest 10% among them was in best agreement with the OMI product: correlation coefficient (0.88), RMSE (1.0%) and mean bias (-0.3%). The 10% average and OMI LER values over ocean were 2% larger in UV than in visible, while the values over land were 1% smaller. The LER variability on the wavelength and surface property was highest (~3%) in the condition of both land and visible, particularly in the ice-cap and desert regions. The minimum reflectance values over the oceanic and inland sample areas overestimated the MODIS product by 1.4%. This high-resolution MODIS observations were effective in removing cloud contamination. The relative errors of the 10% average to MODIS were smaller (-0.6%) over ocean but larger (1.5%) over land than those of the OMI product to MODIS. The reduced relative error in the OMI product over land may result from additional cloud filtering using the Landsat data. This study will be useful when retrieveing the surface reflectance from geostationary-orbit environmental satellite (e.g., Geostationary Environment Monitoring Spectrometer; GEMS).

• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.485-496
• /
• 2022

This study is for the software implementation to generate atmospheric and surface reflectance products from RapidEye satellite imagery. The software is an extension based on Orfeo Toolbox (OTB) and an open-source remote sensing software including calibration modules which use an absolute atmospheric correction algorithm. In order to verify the performance of the program, the accuracy of the product was validated by a test image on the Radiometric Calibration Network (RadCalNet) site. In addition, the accuracy of the surface reflectance product generated from the KOMPSAT-3A image, the surface reflectance of Landsat Analysis Ready Data (ARD) of the same site, and near acquisition date were compared with RapidEye-based one. At the same time, a comparative study was carried out with the processing results using QUick Atmospheric Correction (QUAC) and Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes (FLAASH) tool supported by a commercial tool for the same image. Similar to the KOMPSAT-3A-based surface reflectance product, the results obtained from RapidEye Extension showed accuracy of agreement level within 5%, compared with RadCalNet data. They also showed better accuracy in all band images than the results using QUAC or FLAASH tool. As the importance of the Red-Edge band in agriculture, forests, and the environment applications is being emphasized, it is expected that the utilization of the surface reflectance products of RapidEye images produced using this program will also increase.

• Journal of Korean Society on Water Environment
• /
• v.22 no.5
• /
• pp.815-823
• /
• 2006

The hydrologic cycle in urban catchment has been changed due to the expansion of impervious area by rapid urban development. In this study, the SWMM 5 (Storm Water Management Model 5) model was used to simulate the hydrologic cycle of the Dorimcheon catchment which suffers from the distorted hydrologic cycle as a typical urban catchment. This study compare continuous simulation of urban runoff combining the channel and sewer system with that of channel only in the Dorimcheon catchment. Continuous simulations of urban runoff were performed for the upstream basin of Dorim bridge. The urban impervious regions were processed by the land use analysis from LANDSAT_TM images. It was performed from 1975 to 2000 for every five years. Surface, groundwater and wastewater runoffs were additionally included in the simulations one at a time. Such simulations made it possible to evaluate those components quantitatively. The result of continuous simulation of urban runoff combining the channel and sewer system is that peak flow and recession are well simulated. The analysis results of urbanization effect on runoff are as follows: the surface runoff in 2000 increases to 64% of the whole precipitation whereas the surface runoff in 1975 amounts to 46% of the precipitation; the groundwater runoff in 2000 amounts to 6% and shows 8% decrease during the period from 1975 to 2000.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.04a
• /
• pp.245-250
• /
• 2004

The classification of the satellite images is basic part in Remote sensing. In classification of the satellite images, class separability feature is very effective accuracy of the images classified. For improving classification accuracy, It is necessary to study classification methode than analysis of class separability feature deciding classification probability. In this study, IKONOS, SPOT 5, Landsat TM, were resampled to sizes 1m grid. Above images were calculated the class separability prior to the step for classification of pixels. The results of the study were valued necessary process in geometric information building. This study help to improve accuracy of classification as feature of class separability in the class through optimizing previous classification steps.

• Journal of the Korean Geographical Society
• /
• v.35 no.5
• /
• pp.653-663
• /
• 2000

본 논문에서는 설악산 아고산대에 나타나는 산정형 암괴원에 관한 여러 가지 논의를 제시하였다. 이 암괴원은 산악인들에게는 너덜지대로 너리 알려져 있었으며, 설악산의 북쪽 황철봉과 귀떼기봉에서 대청에 이르는 서북주릉 상 화강암류의 기반암에 암괴원이 분포하고 있다. 이들 암괴원은 규모가 클 뿐망 아니라 암괴상에는 풍화쇄설물을 가지고 있는 나마(gnamma)와 그루브(groove)를 비롯한 많은 화학적 풍화에 의한 미지형들이 발견되고 있어 한반도의 제 4기 후반의 기후변화와 지형형성 영력을 연구할수 이는 기후지형학적으로 중요한 연구지역이라 하겠다. 본 논문에서는 추후 발표될 암괴원의 성인과 고기후학적 의의에 관한 논의에 앞서 먼저 분포지역에 대한 소개와 몇 가지 이론적 고찰에 관하여 논의를 제한하고자 한다. 국내에서는 테일러스, 암괴류 등에 관한 논의가 상당히 이루어지고 있으나 암괴원에 관한 본격적인 논의는 아직 시작단계에 있다. Landsat 영상에 의한 구조선 분석 결과 구조선 밑도는 설악산 지역에서 큰 차이를 나타내지는 않으며, 암괴원은 대부분 화강암 지역에서 나타나는 것으로 판단된다.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1337-1339
• /
• 2003

In this article, a new remote sensing image fusion model based on modified Kohonen networks is given. And a new fusion rule based on modified voting rule was established. Select Shaoxing City as the study site, located at Zhejiang Province, P.R.China. The fusion experiment between Landsat TM data (30m) and IRS-C Pan data (5.8m) was performed using the given fusion method. The fusion results show that the new method can gain better result in apply ing to the lower hill area, and the whole classification accuracy was 10% higher than the basic Kohonen method. The confusion between the woodlands and the waterbodies was also diminished.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2006.05a
• /
• pp.1541-1545
• /
• 2006

본 연구의 목적은 내장저수지의 장래 퇴적량 및 퇴적분포를 추정하는 것으로 연구 결과는 다음과 같다. 1) 내장저수지 유역의 장래 토양유실량을 추정하기 위해 Landsat-5 TM 1984년 영상과 2001년 영상을 분석하여 토지이용변화를 알아보고, 연간 토양유실량 변화를 추정하였다. 2) 내장저수지 유역의 토양유실량을 추정하기 위하여 범용토양유실공식(USLE, Universal Soil Loss Equation)을 이용하였으며, 담수호로 유입되는 양을 추정하기 위하여 유사운송비법을 사용하였다. 담수호로 유입된 유사량 중 퇴적되는 양을 추정하기 위하여 유사량-포착효율법을 이용하였다. 3) 토양유실량의 연평균증가율을 바탕으로 장래 유입 유사량 및 퇴적량을 예측하였으며, Lara법에 의한 수위-내용적 관계 및 표고별 퇴적분포를 추정하였다.

• Proceedings of the Korean Society of Agricultural Engineers Conference
• /
• 2001.10a
• /
• pp.330-334
• /
• 2001

1. The purpose of this study is to evaluate the hydrologic impact due to temporal land cover changes of Gyueongan-cheon watershed. 2. WMS(Watershed Modeling System) HEC-1 was adopted and the required data such as DEM(Digital Elevation Model), stream network, soil map were prepared, and land cover map was made by using Landsat TM data. 3. Due to the increase of urban area and paddy field, the runoff ratio increased 5.8% during the past decade.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.6 no.5
• /
• pp.1-8
• /
• 2003

산림의 절편화 및 마식은 단순한 녹지면직의 축소만이 아니라 아생동물의 이동통로 단절로 인한 서식시 축소 및 니아가서는 소멸을 초래한다. 이를 해결하고자 최근에는 생태통로를 조성하는 작업이 수행되고 있으며 이는 목표종의 신정 및 적절한 위치 선정을 필요로 한다. 본 연구에서는 대전광역시와 그 주변을 대상으로 지난 50년 동안 인간 활동에 의해 절편화된 녹지를 파악하고자 원격탐사(RS) 자료를 이용하였다. Landsat MSS(30m)나 TM(79m)과 같은 중저해상도 위성영상으로는 절편화를 야기하는 산림 절편화 및 마식 파악이 어려워 본연구에서는 발행한 수치지도 및 수치토지이용도를 이용하여 1954년, 2000년 각각의 삼림분포도를 작성하였다. 현지 확인 및 수정과정을 거쳐 최종적으로 생성된 자료를 근거로 산림 절편화 및 마식의 정도를 정량적으로 기술 및 평가하였고 녹지(patch)의 크기와 현장 조사한 야생동물서 식도를 바탕으로 생태통로의 위치를 도출하였다.