• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.83-86
• /
• 2005

This paper presents a method for point-based 3D building reconstruction using LiDAR data and digital map. The proposed method consists of three processes: extraction of building roof points, identification of roof types, and 3D building reconstruction. After extracting points inside the polygon of building, the ground surface, wall and tree points among the extracted points are removed through the filtering process. The filtered points are then fitted into the flat plane using ODR(Orthogonal Distance Regression). If the fitting error is within the predefined threshold, the surface is classified as a flat roof. Otherwise, the surface is fitted and classified into a gable or arch roof through RMSE analysis. Based on the roof types identified in automated fashion, the 3D building reconstruction is performed. Experimental results showed that the proposed method classified successfully three different types of roof and that the fusion of LiDAR data and digital map could be a feasible method of modelling 3D building reconstruction.

• Journal of Aerospace System Engineering
• /
• v.13 no.3
• /
• pp.15-22
• /
• 2019

In this paper, a global and local flight path system for autonomous flight of the UAV is proposed. The overall system is based on the ROS robot operating system. The UAV in-built computer detects obstacles through 2-D Lidar and generates real-time local path and global path based on VFH and Modified $RRT^*$-Smart, respectively. Additionally, a movement command is issued based on the generated path on the UAV flight controller. The ground station computer receives the obstacle information and generates a 2-D SLAM map, transmits the destination point to the embedded computer, and manages the state of the UAV. The autonomous UAV flight system of the is verified through a simulator and actual flight.

• Korean Journal of Remote Sensing
• /
• v.28 no.6
• /
• pp.661-670
• /
• 2012

Urban forests provide great ecosystem services to population in metropolitan areas even though they occupy little green space in a huge gray landscape. Unfortunately, urbanization inherently results in threatening the green infrastructure, and the recent urbanization trends drew great attention of scientists and policy makers on how to preserve or restore green infrastructure in metropolitan area. For this reason, mapping the spatial distribution of the green infrastructure is important in urban environments since the resulting map helps us identify hot green spots and set up long term plan on how to preserve or restore green infrastructure in urban environments. As a preliminary step for mapping green infrastructure utilizing multi-source remote sensing data in urban environments, the objective of this study is to map vegetation volume by fusing LiDAR and multispectral data in urban environments. Multispectral imageries are used to identify the two dimensional distribution of green infrastructure, while LiDAR data are utilized to characterize the vertical structure of the identified green structure. Vegetation volume was calculated over the metropolitan Chicago city area, and the vegetation volume was summarized over 16 NLCD classes. The experimental results indicated that vegetation volume varies greatly even in the same land cover class, and traditional land cover map based above ground biomass estimation approach may introduce bias in the estimation results.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.6
• /
• pp.313-330
• /
• 2021

In an autonomous driving system equipped with a camera and lidar, depth completion techniques enable dense depth estimation. In particular, using self-supervised learning it is possible to train the depth completion network even without ground truth. In actual autonomous driving, such depth completion should have very short latency as it is the input of other algorithms. So, rather than complicate the network structure to increase the accuracy like previous studies, this paper focuses on network latency. We design a U-Net type network with RegNet encoders optimized for GPU computation. Instead, this paper presents several techniques that can increase accuracy during the process of self-supervised learning. The proposed techniques increase the robustness to unreliable lidar inputs. Also, they improve the depth quality for edge and sky regions based on the semantic information extracted in advance. Our experiments confirm that our model is very lightweight (2.42 ms at 1280x480) but resistant to noise and has qualities close to the latest studies.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.4_2
• /
• pp.669-678
• /
• 2023

Recently, spatial information has been applied to various fields and its usability is increasing day by day. In particular, in the field of civil engineering and construction, BIM based on spatial information is being applied to all construction industries and related research has been conducted. BIM is a technology that utilizes spatial information from the design phase and aids in the construction and maintenance of buildings, including the management of their attributes. However, to apply BIM technology to existing buildings, it takes a lot of time and money to produce models based on design drawings along with current surveying. In this study, quantitative and qualitative analysis was conducted to determine the applicability of the acquired data and the applicability of BIM by generating data and analyzing the accuracy using UAV images and ground lidar, which are representative spatial information acquisition methods. Quantitative analysis revealed that TLS (Terrestrial Laser Scanner) showed reliable accuracy in both planar and elevation measurements, whereas unmanned aerial images exhibited lower accuracy in elevation measurements, resulting in reduced reliability. Qualitative analysis indicated that neither TLS nor unmanned aerial images alone provided perfect completeness. However, the combination of both spatial information sources, tailored to specific needs, resulted in the most comprehensive completeness. Therefore, it is concluded that the appropriate utilization of spatial information acquired through unmanned aerial images and TLS holds the potential for application in the fields of BIM and reverse engineering.

• Journal of Korea Multimedia Society
• /
• v.19 no.9
• /
• pp.1647-1658
• /
• 2016

RGB+D database has been widely used in object recognition, object tracking, robot control, to name a few. While rapid advance of active depth sensing technologies allows for the widespread of indoor RGB+D databases, there are only few outdoor RGB+D databases largely due to an inherent limitation of active depth cameras. In this paper, we propose a novel method used to build outdoor RGB+D databases. Instead of using active depth cameras such as Kinect or LIDAR, we acquire a pair of stereo image using high-resolution stereo camera and then obtain a depth map by applying stereo matching algorithm. To deal with estimation errors that inevitably exist in the depth map obtained from stereo matching methods, we develop an approach that estimates confidence of depth maps based on unsupervised learning. Unlike existing confidence estimation approaches, we explicitly consider a spatial correlation that may exist in the confidence map. Specifically, we focus on refining confidence feature with the assumption that the confidence feature and resultant confidence map are smoothly-varying in spatial domain and are highly correlated to each other. Experimental result shows that the proposed method outperforms existing confidence measure based approaches in various benchmark dataset.

• Atmosphere
• /
• v.19 no.4
• /
• pp.319-329
• /
• 2009

The intensive dust observation experiment has been performed at Korea Global Atmosphere Watch Center (KGAW) in Anmyeon, Korea during each spring season from 2007 to 2009. Downward and upward hyper-spectral spectrums over the dust condition were measured to understand the hyper-spectral properties of Asian dust using both ground-based Fourier Transform Infrared Spectroscopy (FT-IR) and space-borne AIRS/Aqua. To understand the impact of the Asian dust, a Line-by-Line radiative transfer model runs to calculate the high resolution infrared spectrum over the wave number range of $500-500cm^{-1}$. Furthermore, the radiosonde, a $PM_{10}$ Sampler, a Micro Pulse Lidar (MPL), and an Aerodynamic Particle Sizer (APS) are used to understand the vertical profile of temperature and humidity and the properties of Asian dust like concentration, altitude of dust layer, and size distribution. In this study, we found the Asian dust distributed from surface up to 3-4 km and volume concentration is increased at the size range between 2 and $8{\mu}m$ The observed dust spectrums are larger than the calculated clear sky spectrums by 15~60K for downward and lower by around 2~6K for upward in the wave number range of $800-1200cm^{-1}$. For the characteristics of the spectrum during the Asian dust, the downward spectrum is revealed a positive slope for $800-1000cm^{-1}$ region and negative slope over $1100-1200cm^{-1}$ region. In the upward spectrum, slopes are opposed to the downward one. It is inferred that the difference between measured and calculated spectrum is mostly due to the contribution of emission and/or absorption of the dust particles by the aerosol amount, size distribution, altitude, and composition.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2006.04a
• /
• pp.205-210
• /
• 2006

With the development of recent technology such as telematics, LBS, and ubiquitous, the applications of 3D GIS are rapidly increased. As 3D GIS is mainly based on urban models consisting of the realistic digital models of the objects existing in an urban area, demands for urban models and its continuous update is expected to be drastically increased. The purpose of this study is thus to propose more efficient and precise methods to construct urban models with its experimental verification. Applying the proposed methods, the terrain and sophisticated building models are constructed for the area of $270,600m^2$ with 23 buildings in the University of Seoul. For the terrain models, airborne imagery and LIDAR data is used, while the ground imagery is mainly used for the building models. It is found that the generated models reflect the correct geometry of the buildings and terrain surface. The textures of building surfaces, generated automatically using the projective transformation however, are not well-constructed because of being blotted out and shaded by objects such as trees, near buildings, and other obstacles. Consequently, the algorithms on the texture extraction should be improved to construct more realistic 3D models. Furthermore, the inside of buildings should be modeled for various potential applications in the future.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.6
• /
• pp.545-554
• /
• 2019

This research aims to provide a building area extraction approach over the areas where data acquisition is impossible through field surveying, aerial photography and lidar scanning. Hence, high-resolution satellite images, which have high accessibility over the earth, are utilized for the automated building extraction in this study. 3D point clouds or DSM (Digital Surface Models), derived from the stereo image matching process, provides low quality of building area extraction due to their high level of noises and holes. In this regards, this research proposes a hybrid building area extraction approach which utilizes 3D point clouds (from image matching), and color and linear information (from imagery). First of all, ground and non-ground points are separated from 3D point clouds; then, the initial building hypothesis is extracted from the non-ground points. Secondly, color based building hypothesis is produced by considering the overlapping between the initial building hypothesis and the color segmentation result. Afterwards, line detection and space partitioning results are utilized to acquire the final building areas. The proposed approach shows 98.44% of correctness, 95.05% of completeness, and 1.05m of positional accuracy. Moreover, we see the possibility that the irregular shapes of building areas can be extracted through the proposed approach.

• Atmosphere
• /
• v.18 no.4
• /
• pp.267-277
• /
• 2008

The occurrence and strength of the Asian Dust over the Korea Peninsular have been increased by the expansion of the desert area. For the continuous monitoring of the Asian Dust event, the geostationary satellites provide useful information by detecting the outbreak of the event as well as the long-range transportation of dust. The Infrared Optical Depth Index (IODI) derived from the MTSAT-1R data, indicating a quantitative index of the dust intensity, has been produced in real-time at Korea Meteorological Administration (KMA) since spring of 2007 for the forecast of Asian dust. The data processing algorithm for IODI consists of mainly two steps. The first step is to detect dust area by using brightness temperature difference between two thermal window channels which are influenced with different extinction coefficients by dust. Here we use dynamic threshold values based on the change of surface temperature. In the second step, the IODI is calculated using the ratio between current IR1 brightness temperature and the maximum brightness temperature of the last 10 days which we assume the clear sky. Validation with AOD retrieved from MODIS shows a good agreement over the ocean. Comparison of IODI with the ground based PM10 observation network in Korea shows distinct characteristics depending on the altitude of dust layer estimated from the Lidar data. In the case that the altitude of dust layer is relatively high, the intensity of IODI is larger than that of PM10. On the other hand, when the altitude of dust layer is lower, IODI seems to be relatively small comparing with PM10 measurement.