• Title/Summary/Keyword: 3D PointCloud

Search Result 410, Processing Time 0.03 seconds

Classification of 3D Road Objects Using Machine Learning (머신러닝을 이용한 3차원 도로객체의 분류)

  • Hong, Song Pyo;Kim, Eui Myoung
    • Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
    • /
    • v.36 no.6
    • /
    • pp.535-544
    • /
    • 2018
  • Autonomous driving can be limited by only using sensors if the sensor is blocked by sudden changes in surrounding environments or large features such as heavy vehicles. In order to overcome the limitations, the precise road-map has been used additionally. This study was conducted to segment and classify road objects using 3D point cloud data acquired by terrestrial mobile mapping system provided by National Geographic Information Institute. For this study, the original 3D point cloud data were pre-processed and a filtering technique was selected to separate the ground and non-ground points. In addition, the road objects corresponding to the lanes, the street lights, the safety fences were initially segmented, and then the objects were classified using the support vector machine which is a kind of machine learning. For the training data for supervised classification, only the geometric elements and the height information using the eigenvalues extracted from the road objects were used. The overall accuracy of the classification results was 87% and the kappa coefficient was 0.795. It is expected that classification accuracy will be increased if various classification items are added not only geometric elements for classifying road objects in the future.

Implementation of File-referring Octree for Huge 3D Point Clouds (대용량 3차원 포인트 클라우드를 위한 파일참조 옥트리의 구현)

  • Han, Soohee
    • Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
    • /
    • v.32 no.2
    • /
    • pp.109-115
    • /
    • 2014
  • The aim of the study is to present a method to build an octree and to query from it for huge 3D point clouds of which volumes correspond or surpass the main memory, based on the memory-efficient octree developed by Han(2013). To the end, the method directly refers to 3D point cloud stored in a file on a hard disk drive instead of referring to that duplicated in the main memory. In addition, the method can save time to rebuild octree by storing and restoring it from a file. The memory-referring method and the present file-referring one are analyzed using a dataset composed of 18 million points surveyed in a tunnel. In results, the memory-referring method enormously exceeded the speed of the file-referring one when generating octree and querying points. Meanwhile, it is remarkable that a still bigger dataset composed of over 300 million points could be queried by the file-referring method, which would not be possible by the memory-referring one, though an optimal octree destination level could not be reached. Furthermore, the octree rebuilding method proved itself to be very efficient by diminishing the restoration time to about 3% of the generation time.

A Real-time Plane Estimation in Virtual Reality Using a RGB-D Camera in Indoors (RGB-D 카메라를 이용한 실시간 가상 현실 평면 추정)

  • Yi, Chuho;Cho, Jungwon
    • Journal of Digital Convergence
    • /
    • v.14 no.11
    • /
    • pp.319-324
    • /
    • 2016
  • In the case of robot and Argument Reality applications using a camera in environments, a technology to estimate planes is a very important technology. A RGB-D camera can get a three-dimensional measurement data even in a flat which has no information of the texture of the plane;, however, there is an enormous amount of computation in order to process the point-cloud data of the image. Furthermore, it could not know the number of planes that are currently observed as an advance, also, there is an additional operation required to estimate a three dimensional plane. In this paper, we proposed the real-time method that decides the number of planes automatically and estimates the three dimensional plane by using the continuous data of an RGB-D camera. As experimental results, the proposed method showed an improvement of approximately 22 times faster speed compared to processing the entire data.

Valve Model Extraction from Noisy 3-D Point Cloud Data (잡음이 있는 3차원 점군 정보에서의 밸브 모형 추출)

  • Oh, Ki-Won;Choi, Kang-Sun
    • Annual Conference of KIPS
    • /
    • 2015.04a
    • /
    • pp.945-946
    • /
    • 2015
  • Laser Range Finder를 이용하여 생성한 3차원 점군 정보는 단면적인 부분만 볼 수 있으며, 잡음이 포함되어 작은 물체를 추출하는데 많은 영향이 생긴다. 이러한 잡음이 있는 3차원 점군 정보 사이에서 밸브의 중심의 위치에 대한 추가적인 입력을 받아 원환체, 원통, 평면의 정보를 복합적으로 포함하고 있는 밸브의 모델을 추출한다.

An Efficient Polygonal Surface Reconstruction (효율적인 폴리곤 곡면 재건 알고리즘)

  • Park, Sangkun
    • Journal of Institute of Convergence Technology
    • /
    • v.10 no.1
    • /
    • pp.7-12
    • /
    • 2020
  • We describe a efficient surface reconstruction method that reconstructs a 3D manifold polygonal mesh approximately passing through a set of 3D oriented points. Our algorithm includes 3D convex hull, octree data structure, signed distance function (SDF), and marching cubes. The 3D convex hull provides us with a fast computation of SDF, octree structure allows us to compute a minimal distance for SDF, and marching cubes lead to iso-surface generation with SDF. Our approach gives us flexibility in the choice of the resolution of the reconstructed surface, and it also enables to use on low-level PCs with minimal peak memory usage. Experimenting with publicly available scan data shows that we can reconstruct a polygonal mesh from point cloud of sizes varying from 10,000 ~ 1,000,000 in about 1~60 seconds.

Development of Digital Surface Model and Feature Extraction by Integrating Laser Scanner and CCD sensor

  • Nagai, Masahiko;Shibasaki, Ryosuke;Zhao, Huijing;Manandhar, Dinesh
    • Proceedings of the KSRS Conference
    • /
    • 2003.11a
    • /
    • pp.859-861
    • /
    • 2003
  • In order to present a space in details, it is indispensable to acquire 3D shape and texture simultaneously from the same platform. 3D shape is acquired by Laser Scanner as point cloud data, and texture is acquired by CCD sensor. Positioning data is acquired by IMU (Inertial Measurement Unit). All the sensors and equipments are assembled on a hand-trolley. In this research, a method of integrating the 3D shape and texture for automated construction of Digital Surface Model is developed. This Digital Surface Model is applied for efficient feature extraction. More detailed extraction is possible , because 3D Digital Surface Model has both 3D shape and texture information.

  • PDF

3D Point Cloud Compression Using Spatial Distribution Information (공간적 분포 정보를 이용한 3D 포인트 클라우드 압축)

  • Kim, Ji-Su;Lee, Seonho;Lee, Se-Ho;Kim, Chang-Su
    • Proceedings of the Korean Society of Broadcast Engineers Conference
    • /
    • 2018.06a
    • /
    • pp.221-222
    • /
    • 2018
  • 본 논문에서는 공간적 분포 정보를 이용한 3D 포인트 클라우드 압축 기법을 제안한다. 우선, 3D 포인트 클라우드에 대해 팔진 트리 구조를 생성한다. 그리고 잎사귀 노드들에 대해서 해당 복셀의 중심으로부터의 유클리드 거리를 구하고, 이를 통해서 공간적 분포 정보를 구성한다. 이때, 복셀 내 포인트들의 분포를 고려하여, 포인트들이 밀집하여 분포하는 경우 복셀 내 포인트들을 하나의 대표 위치로 표현하여 압축한다. 최종적으로 산술부호화를 수행하여 압축 성능을 높인다. 실험 결과 제안하는 기법이 3D 포인트 클라우드를 효율적으로 압축함을 확인할 수 있다.

  • PDF

Selection of Transition Point through Calculation of Cumulative Toxic Load -Focused on Incheon Area- (누적독성부하 산정을 통한 주민소산 전환시점 선정에 관한 연구 -인천지역을 중심으로-)

  • Lee, Eun Ji;Han, Man Hyeong;Chon, Young Woo;Lee, Ik Mo;Hwang, Yong Woo
    • Journal of the Korean Society of Safety
    • /
    • v.35 no.6
    • /
    • pp.15-24
    • /
    • 2020
  • With the development of the chemical industry, the chemical accident is increasing every year, thereby increasing the risk of accidents caused by chemicals. The Ministry of Environment provides the criteria for determining shelter-in-place or outdoor evacuation by material, duration of accident, and distance from the toxic substance leak. However, it is hard to say that the criteria for determining the transition point are not clear. Transition point mean the time that evacuation method is switched from shelter-in-place to outdoor evacuation. So, the purpose of this study was to calculate appropriate transition point by comparing the cumulative toxic load. Namdong-gu in Incheon Metropolitan City was finally selected as the target area, considering the current status of the population of Incheon Metropolitan City in 2016 and the statistical survey of chemicals in 2016. The target materials were HCl, HF, and NH3. Modeling was simulated by ALOHA and performed assuming that the entire amount would be leaked for 10 min. Residents' evacuation scenarios were assumed to be shelter-in-place, immediate outdoor evacuation, and outdoor evacuation at an appropriate time after shelter-in-place. Based on the above method, the appropriate transition point from residents located in A(800 m away), B(1,200 m away), C(1,400 m away) and D(2,200 m away) was identified. In HCl, appropriate transition point was after 15 min, after 16 min, after 17 min, after 20 min in order by A, B, C and D. In HF, appropriate transition point was before 1 min or after 16 min, before 4 min or after 19 min, before 5 min or after 20 min, before 14 min or after 26 min in order by A, B, C and D. In NH3, appropriate transition point at A was before 4 min or after 16. Others are not in chemical cloud. This study confirmed the transition point to minimize the cumulative toxic load can be obtained by quantitative method. Through this, it might be possible to select evacuation method quantitatively that cumulative toxic load are minimal. In addition, if the shelter-in-place is maintained without transition to outdoor evacuation, the cumulative toxic load will increase more than outdoor evacuation. Therefore, it was confirmed that actions to reduce the concentration of chemicals in the room were necessary, such as conducting ventilation after the chemical cloud passed through the site.

MPEG G-PCC 국제표준 기술

  • Byeon, Ju-Hyeong;Choe, Han-Sol;Sim, Dong-Gyu
    • Broadcasting and Media Magazine
    • /
    • v.26 no.2
    • /
    • pp.31-45
    • /
    • 2021
  • 본 고는 ISO/IEC JTC 1/SC 29/WG 7 MPEG(Moving Picture Experts Group) 3DG(3D Graphics coding) 그룹에서 진행되고 있는 포인트 클라우드 데이터 압축 표준 기술 중 하나인 G-PCC(Geometry-based Point Cloud Compression) 표준에 대하여 설명하고자 한다. G-PCC는 포인트 클라우드의 기하 정보와 속성 정보를 3차원 공간에서 서로 다른 기술을 이용하여 압축하는 표준으로, 무손실 압축 방법의 경우 10:1의 압축율을 제공하고 손실 압축의 경우 35:1 정도의 압축율을 보인다. 본 고에서는 G-PCC의 기하 정보와 속성 정보의 압축 방법을 상세히 설명하고 같은 기능을 수행하는 압축 기술 간의 성능을 비교하고자 한다.

Featured-Based Registration of Terrestrial Laser Scans with Minimum Overlap Using Photogrammetric Data

  • Renaudin, Erwan;Habib, Ayman;Kersting, Ana Paula
    • ETRI Journal
    • /
    • v.33 no.4
    • /
    • pp.517-527
    • /
    • 2011
  • Currently, there is a considerable interest in 3D object reconstruction using terrestrial laser scanner (TLS) systems due to their ability to automatically generate a considerable amount of points in a very short time. To fully map an object, multiple scans are captured. The different scans need to be registered with the help of the point cloud in the overlap regions. To guarantee reliable registration, the scans should have large overlap ratio with good geometry for the estimation of the transformation parameters among these scans. The objective of this paper is to propose a registration method that relaxes/eliminates the overlap requirement through the utilization of photogrammetrically reconstructed features. More specifically, a point-based procedure, which utilizes non-conjugate points along corresponding linear features from photogrammetric and TLS data, will be used for the registration. The non-correspondence of the selected points along the linear features is compensated for by artificially modifying their weight matrices. The paper presents experimental results from simulated and real datasets to illustrate the feasibility of the proposed procedure.