항공 LiDAR 데이터를 이용한 건물추출과 상부구조물 특성분석 및 모델링

정형섭(Jung, Hyung Sup),임새봄(Lim, Sae Bom),이동천(Lee, Dong Cheon) 한국측량학회 2008 한국측량학회지 Vol.26 No.3

SPOT 4 영상의 기하보정을 위한 시선 벡터 조정 모델

정형섭(Jung, Hyung-Sup) 한국측량학회 2010 한국측량학회지 Vol.28 No.2

차분 간섭도로부터 지표변위의 시계열 관측을 위한 개선된 Small Baseline Subset (SBAS) 알고리즘

정형섭 ( Hyung Sup Jung ),이창욱 ( Chang Wook Lee ),박정원 ( Jung Won Park ),김기동 ( Ki Dong Kim ),원중선 ( Joong Sun Won ) 大韓遠隔探査學會 2008 大韓遠隔探査學會誌 Vol.24 No.2

최근 spatial decorrelation을 극복하기 위하여 비교적 짧은 기선(baseline)을 지니는 여러 시기의 차분간섭도(differential interferogram)로부터 시계열 지표변위(time-series surface deformation)를 관측할 수 있는 small baseline subset(SBAS) 알고리즘이 개발되었다. 이 SBAS 알고리즘은 singluar value decomposition(SVD)을 이용하여 시간별로 완벽하게 연결되지 않는 차분간섭도로부터 시계열 지표변위를 관측하였을 뿐 아니라 공간적으로 저주파 필터와 시간적으로 고주파 필터를 이용하여 대기효과를 보정하였다. 그러나 이 알고리즘은 초기 관측시 시계열 지표변위를 선형으로 가정하였기 때문에 각 차분간섭도의 phase unwrapping 오차를 정확하게 보정하기 어려웠을 뿐 아니라 시계열의 지표변위에 존재하는 노이즈 성분을 완화시키지 못했다. 이와 같은 단점을 보완하기 위하여 이 연구에서는 기존의 SBAS 알고리즘을 개선하였다. 이 개선된 SBAS 알고리즘은 각 차분간섭도의 phase unwrapping 오차를 최소화하기 위하여 반복적으로 시계열 지표면 변위를 개선하였고, 시계열 지표변위의 노이즈를 제거하기 위하여 유한차분 근사법(finite difference approximation)을 이용하였다. 서로 다른 지역의 26개의 ERS-1/2자료와 21개의 RADARSAT-1 fine beam (F5) 자료를 이용하여 개선된 SBAS 알고리즘을 실험하고 분석하였다. ERS-1/2자료에서는 LOS(line-of-sight) 지표변위가 약 13년 동안 최대 -40cm가 관측되었고, RADARSAT-1 fine beam 자료에서는 약 2년 동안 최대 -3cm의 LOS 지표변위가 관측되었다. Small baseline subset (SBAS) algorithm has been recently developed using an appropriate combination of differential interferograms, which are characterized by a small baseline in order to minimize the spatial decorrelation. This algorithm uses the singular value decomposition (SVD) to measure the time-series surface deformation from the differential interferograms which are not temporally connected. And it mitigates the atmospheric effect in the time-series surface deformation by using spatially low-pass and temporally high-pass filter. Nevertheless, it is not easy to correct the phase unwrapping error of each interferogram and to mitigate the time-varying noise component of the surface deformation from this algorithm due to the assumption of the linear surface deformation in the beginning of the observation. In this paper, we present an improved SBAS technique to complement these problems. Our improved SBAS algorithm uses an iterative approach to minimize the phase unwrapping error of each differential interferogram. This algorithm also uses finite difference method to suppress the time-varying noise component of the surface deformation. We tested our improved SBAS algorithm and evaluated its performance using 26 images of ERS-1/2 data and 21 images of RADARSAT-1 fine beam (F5) data at each different locations. Maximum deformation amount of 40cm in the radar line of sight (LOS) was estimated from ERS-1/2 datasets during about 13 years, whereas 3cm deformation was estimated from RADARSAT-1 ones during about two years.

항공영상으로부터 YOLOv5를 이용한 도심수목 탐지

박채원,정형섭,Park, Che-Won,Jung, Hyung-Sup 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6

Urban population concentration and indiscriminate development are causing various environmental problems such as air pollution and heat island phenomena, and causing human resources to deteriorate the damage caused by natural disasters. Urban trees have been proposed as a solution to these urban problems, and actually play an important role, such as providing environmental improvement functions. Accordingly, quantitative measurement and analysis of individual trees in urban trees are required to understand the effect of trees on the urban environment. However, the complexity and diversity of urban trees have a problem of lowering the accuracy of single tree detection. Therefore, we conducted a study to effectively detect trees in Dongjak-gu using high-resolution aerial images that enable effective detection of tree objects and You Only Look Once Version 5 (YOLOv5), which showed excellent performance in object detection. Labeling guidelines for the construction of tree AI learning datasets were generated, and box annotation was performed on Dongjak-gu trees based on this. We tested various scale YOLOv5 models from the constructed dataset and adopted the optimal model to perform more efficient urban tree detection, resulting in significant results of mean Average Precision (mAP) 0.663.

딥러닝 기반 레이더 간섭 위상 언래핑 기술 고찰

백원경,정형섭,Baek, Won-Kyung,Jung, Hyung-Sup 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6

위상 언래핑은 위성레이더 간섭기법의 필수적인 자료처리 절차다. 이에 따라 비 딥러닝 기반 언래핑 기법이 다수 개발되었으며 최근에는 딥러닝 기반 언래핑 기법이 제안되고 있다. 본 논문에서는 딥러닝 기반 위성레이더 언래핑 기법을 1) 언래핑된 위상의 예측 방법, 2) 위상 언래핑을 위한 딥러닝 모델의 구조 그리고 3) 학습데이터 제작 방법의 측면에서 최근 연구 동향을 소개하였다. 언래핑된 위상을 예측하는 방법은 모호 정수 분류방법, 위상 단절 구간 탐지 방법, 위상 예측 방법, 딥러닝과 전통적인 언래핑 기법의 연계 방법에 따라 다시 세분화하여 연구 동향을 나타냈다. 일반적으로 활용되는 딥러닝 모델 구조의 특징과 전체 위상 정보를 파악하기 위한 모델 최적화 방법에 대한 연구 사례를 소개하였다. 또한 학습데이터 제작 방법은 주로 위상 변이 제작과 노이즈 시뮬레이션 방법으로 구분하여 연구 동향을 정리하였으며 추후 발전 방향을 제시하였다. 본 논문이 추후 국내의 딥러닝 기반 위상 언래핑 연구의 발전 방향을 모색하는 데에 필요한 기반 자료로 활용되기를 기대한다.

고해상도 원격탐사 영상을 이용한 YOLOv5기반 굴뚝 탐지

윤영웅,정형섭,이원진,Yoon, Young-Woong,Jung, Hyung-Sup,Lee, Won-Jin 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6

Air pollution is social issue that has long-term and short-term harmful effect on the health of animals, plants, and environments. Chimneys are the primary source of air pollutants that pollute the atmosphere, so their location and type must be detected and monitored. Power plants and industrial complexes where chimneys emit air pollutants, are much less accessible and have a large site, making direct monitoring cost-inefficient and time-inefficient. As a result, research on detecting chimneys using remote sensing data has recently been conducted. In this study, YOLOv5-based chimney detection model was generated using BUAA-FFPP60 open dataset create for power plants in Hebei Province, Tianjin, and Beijing, China. To improve the detection model's performance, data split and data augmentation techniques were used, and a training strategy was developed for optimal model generation. The model's performance was confirmed using various indicators such as precision and recall, and the model's performance was finally evaluated by comparing it to existing studies using the same dataset.

항공정사영상의 상대적인 지상좌표 위치오차에 따른 색상보정

박숭환,정형섭,정경식,김경휘,Park, Sung-Hwan,Jung, Hyung-Sup,Jung, Kyungsik,Kim, Kyong-Hwi 대한원격탐사학회 2017 大韓遠隔探査學會誌 Vol.33 No.5

본 연구는 동일 촬영시기 정사영상에 대한 색상보정을 효과적으로 수행하기 위하여 실시되었다. 이를 위하여, 인접 영상 간 상대적인 지상좌표 위치오차를 분석하였다. 위치오차를 저감시키기 위한 방법으로 block sum 방법을 제시하였다. 각각 block sum 크기에 따라 결정된 회귀계수를 이용하여 상대색상보정을 수행하였다. 그 결과 시각적으로 상대색상보정이 잘 수행되었음을 확인하였다. 정량적인 분석은 히스토그램 유사성 분석을 통해 수행되었다. 이로부터 block sum 방법이 상대색상보정에 유용함이 증명되었다. 특히 상대적인 지상좌표 위치오차의 양에 따라 block sum 크기의 선정이 매우 중요한 것으로 나타났다. 위치오차가 클수록 높은 크기에서의 block sum 적용이 유용한 것으로 확인되었다. This study was carried out to effectively perform relative color correction for high-resolution aerial ortho image. For this study, relative geometrical error between adjacent images was analyzed. The block sum method is proposed to reduce the relative geometrical error. We used the regression coefficients determined based on the block sum size to perform the color correction. As a result, it was confirmed that the relative color correction was visually performed well. Quantitative analysis was performed through histogram similarity analysis. It is proved that block sum method is useful for relative color correction. Particularly, the block sum size was very important to correct color based on the amount of relative geometrical error.

시계열 위성영상을 위한 효과적인 Super Resolution 기법

정승균,최윤수,정형섭,Jung, Seung-Kyoon,Choi, Yun-Soo,Jung, Hyung-Sup 한국공간정보학회 2011 한국공간정보학회지 Vol.19 No.1

정지궤도 해색 센서(GOCI: Geostationary Ocean Color Imager) 는 세계 최초의 정지궤도 위성으로 매일 1 시간마다 8 장의 영상을 획득 할 수 있어 육상파 해양 모두 활용성이 높은 위성이다. 그러나 500m의 GSD(Ground Sample Distance)를 지니는 서해성도 영상은 육성 활용에 한계가 있다. 최근, 컴퓨터 비전분야에서 활발히 진행 중인 기술인 Super Resolution(이하 SR)는 유사 시간대에 촬영한 저해상도 영상으로부터 고해상도 영상을 제작하는 기술로, 이를 시간 해상도가 높은 시계열 위성인 GOCI에 적용한다면 해상도가 향상 된 영상을 제작하는 기술로, 이를 시간 해상도가 높은 시계열 위성인 GOCI에 적용한다면 해상도가 향상 된 영상의 취득이 가능하며, 또한 광학 위성 영상의 단점인 구름에 의해 손실된 지상 정보의 복원이 가능할 것이다. 본 연구에서는, GOCI 자료를 위한 효율적인 초해상도 영상 복원 알고리즘 개발을 위한 선행연구로써 위성 영상 취득과정과 유사한 환경의 시뮬레이션을 통해 시계열 자료를 제작하고, 제작된 자료를 제안한 알고리즘에 적용함으로서 0.1 단위의 픽셀 정합도를 확인하였고, 원본 영상과 RMSE 0.5763, PSNR 52.9183 db, SSIM Index 0.9486의 정확도를 나타낸 HR 영상을 복원하였다. GOCI the world first Ocean Color Imager in Geostationary Orbit, which could obtain total 8 images of the same region a day, however, its spatial resolution(500m) is not enough to use for the accurate land application, Super Resolution(SR), reconstructing the high resolution(HR) image from multiple low resolution(LR) images introduced by computer vision field. could be applied to the time-series remotely sensed images such as GOCI data, and the higher resolution image could be reconstructed from multiple images by the SR, and also the cloud masked area of images could be recovered. As the precedent study for developing the efficient SR method for GOCI images, on this research, it reproduced the simulated data under the acquisition process of the remote sensed data, and then the simulated images arc applied to the proposed algorithm. From the proposed algorithm result of the simulated data, it turned out that low resolution(LR) images could be registered in sub-pixel accuracy, and the reconstructed HR image including RMSE, PSNR, SSIM Index value compared with original HR image were 0.5763, 52.9183 db, 0.9486, could be obtained.

데이터 확장을 통한 토지피복분류 U-Net 모델의 성능 개선

백원경,이명진,정형섭,Baek, Won-Kyung,Lee, Moung-Jin,Jung, Hyung-Sup 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6

Recently, a number of deep-learning based land cover segmentation studies have been introduced. Some studies denoted that the performance of land cover segmentation deteriorated due to insufficient training data. In this study, we verified the improvement of land cover segmentation performance through data augmentation. U-Net was implemented for the segmentation model. And 2020 satellite-derived landcover dataset was utilized for the study data. The pixel accuracies were 0.905 and 0.923 for U-Net trained by original and augmented data respectively. And the mean F1 scores of those models were 0.720 and 0.775 respectively, indicating the better performance of data augmentation. In addition, F1 scores for building, road, paddy field, upland field, forest, and unclassified area class were 0.770, 0.568, 0.433, 0.455, 0.964, and 0.830 for the U-Net trained by original data. It is verified that data augmentation is effective in that the F1 scores of every class were improved to 0.838, 0.660, 0.791, 0.530, 0.969, and 0.860 respectively. Although, we applied data augmentation without considering class balances, we find that data augmentation can mitigate biased segmentation performance caused by data imbalance problems from the comparisons between the performances of two models. It is expected that this study would help to prove the importance and effectiveness of data augmentation in various image processing fields.

Deep Neural Network와 Convolutional Neural Network 모델을 이용한 산사태 취약성 매핑

공성현,백원경,정형섭,Gong, Sung-Hyun,Baek, Won-Kyung,Jung, Hyung-Sup 대한원격탐사학회 2022 大韓遠隔探査學會誌 Vol.38 No.6

Landslides are one of the most prevalent natural disasters, threating both humans and property. Also landslides can cause damage at the national level, so effective prediction and prevention are essential. Research to produce a landslide susceptibility map with high accuracy is steadily being conducted, and various models have been applied to landslide susceptibility analysis. Pixel-based machine learning models such as frequency ratio models, logistic regression models, ensembles models, and Artificial Neural Networks have been mainly applied. Recent studies have shown that the kernel-based convolutional neural network (CNN) technique is effective and that the spatial characteristics of input data have a significant effect on the accuracy of landslide susceptibility mapping. For this reason, the purpose of this study is to analyze landslide vulnerability using a pixel-based deep neural network model and a patch-based convolutional neural network model. The research area was set up in Gangwon-do, including Inje, Gangneung, and Pyeongchang, where landslides occurred frequently and damaged. Landslide-related factors include slope, curvature, stream power index (SPI), topographic wetness index (TWI), topographic position index (TPI), timber diameter, timber age, lithology, land use, soil depth, soil parent material, lineament density, fault density, normalized difference vegetation index (NDVI) and normalized difference water index (NDWI) were used. Landslide-related factors were built into a spatial database through data preprocessing, and landslide susceptibility map was predicted using deep neural network (DNN) and CNN models. The model and landslide susceptibility map were verified through average precision (AP) and root mean square errors (RMSE), and as a result of the verification, the patch-based CNN model showed 3.4% improved performance compared to the pixel-based DNN model. The results of this study can be used to predict landslides and are expected to serve as a scientific basis for establishing land use policies and landslide management policies.