• Journal of the Society of Naval Architects of Korea
• /
• v.59 no.5
• /
• pp.330-337
• /
• 2022

Research on object detection algorithms using 2D data has already progressed to the level of commercialization and is being applied to various manufacturing industries. Object detection technology using 2D data has an effective advantage, there are technical limitations to accurate data generation and analysis. Since 2D data is two-axis data without a sense of depth, ambiguity arises when approached from a practical point of view. Advanced countries such as the United States are leading 3D data collection and research using 3D laser scanners. Existing processing and detection algorithms such as ICP and RANSAC show high accuracy, but are used as a processing speed problem in the processing of large-scale point cloud data. In this study, PointNet a representative technique for detecting objects using widely used 3D point cloud data is analyzed and described. And RandLA-Net, which overcomes the limitations of PointNet's performance and object prediction accuracy, is described a review of detection technology using point cloud data was conducted.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.528-529
• /
• 2023

Due to vast points and irregular structure, labeling full points in large-scale point clouds is highly tedious and time-consuming. To resolve this issue, we propose a novel point-based transformer network in weakly-supervised semantic segmentation, which only needs 0.1% point annotations. Our network introduces general local features, representing global factors from different neighborhoods based on their order positions. Then, we share query point weights to local features through point attention to reinforce impacts, which are essential in determining sparse point labels. Geometric encoding is introduced to balance query point impact and remind point position during training. As a result, one point in specific local areas can obtain global features from corresponding ones in other neighborhoods and reinforce from its query points. Experimental results on benchmark large-scale point clouds demonstrate our proposed network's state-of-the-art performance.

• The Journal of Korea Robotics Society
• /
• v.10 no.1
• /
• pp.33-41
• /
• 2015

Recent developments in robotics and intelligent vehicle area, bring interests of people in an autonomous driving ability and advanced driving assistance system. Especially fully automatic parking ability is one of the key issues of intelligent vehicles, and accurate parked vehicles detection is essential for this issue. In previous researches, many types of sensors are used for detecting vehicles, 2D LiDAR is popular since it offers accurate range information without preprocessing. The L shape feature is most popular 2D feature for vehicle detection, however it has an ambiguity on different objects such as building, bushes and this occurs misdetection problem. Therefore we propose the accurate vehicle detection method by using a 3D complete vehicle model in 3D point clouds acquired from front inclined 2D LiDAR. The proposed method is decomposed into two steps: vehicle candidate extraction, vehicle detection. By combination of L shape feature and point clouds segmentation, we extract the objects which are highly related to vehicles and apply 3D model to detect vehicles accurately. The method guarantees high detection performance and gives plentiful information for autonomous parking. To evaluate the method, we use various parking situation in complex urban scene data. Experimental results shows the qualitative and quantitative performance efficiently.

• ETRI Journal
• /
• v.37 no.3
• /
• pp.606-616
• /
• 2015

In this paper, we propose a novel method for road recognition using 3D point clouds based on a Markov random field (MRF) framework in unstructured and complex road environments. The proposed method is focused on finding a solution for an analysis of traversable regions in challenging environments without considering an assumption that has been applied in many past studies; that is, that the surface of a road is ideally flat. The main contributions of this research are as follows: (a) guidelines for the best selection of the gradient value, the average height, the normal vectors, and the intensity value and (b) how to mathematically transform a road recognition problem into a classification problem that is based on MRF modeling in spatial and visual contexts. In our experiments, we used numerous scans acquired by an HDL-64E sensor mounted on an experimental vehicle. The results show that the proposed method is more robust and reliable than a conventional approach based on a quantity evaluation with ground truth data for a variety of challenging environments.

• Journal of Computational Design and Engineering
• /
• v.1 no.1
• /
• pp.13-26
• /
• 2014

Recently, renovations of plant equipment have been more frequent because of the shortened lifespans of the products, and as-built models from large-scale laser-scanned data is expected to streamline rebuilding processes. However, the laser-scanned data of an existing plant has an enormous amount of points, captures intricate objects, and includes a high noise level, so the manual reconstruction of a 3D model is very time-consuming and costly. Among plant equipment, piping systems account for the greatest proportion. Therefore, the purpose of this research was to propose an algorithm which could automatically recognize a piping system from the terrestrial laser-scanned data of plant equipment. The straight portion of pipes, connecting parts, and connection relationship of the piping system can be recognized in this algorithm. Normal-based region growing and cylinder surface fitting can extract all possible locations of pipes, including straight pipes, elbows, and junctions. Tracing the axes of a piping system enables the recognition of the positions of these elements and their connection relationship. Using only point clouds, the recognition algorithm can be performed in a fully automatic way. The algorithm was applied to large-scale scanned data of an oil rig and a chemical plant. Recognition rates of about 86%, 88%, and 71% were achieved straight pipes, elbows, and junctions, respectively.

• Journal of Computational Design and Engineering
• /
• v.3 no.2
• /
• pp.102-111
• /
• 2016

In this paper, we study the problem of computing smooth feature curves from CAD type point clouds models. The proposed method reconstructs feature curves from the intersections of developable strip pairs which approximate the regions along both sides of the features. The generation of developable surfaces is based on a linear approximation of the given point cloud through a variational shape approximation approach. A line segment sequencing algorithm is proposed for collecting feature line segments into different feature sequences as well as sequential groups of data points. A developable surface approximation procedure is employed to refine incident approximation planes of data points into developable strips. Some experimental results are included to demonstrate the performance of the proposed method.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.792-799
• /
• 2022

Building Information Modeling (BIM) technology is a key component of modern construction engineering and project management workflows. As-is BIM models that represent the spatial reality of a project site can offer crucial information to stakeholders for construction progress monitoring, error checking, and building maintenance purposes. Geometric methods for automatically converting raw scan data into BIM models (Scan-to-BIM) often fail to make use of higher-level semantic information in the data. Whereas, semantic segmentation methods only output labels at the point level without creating object level models that is necessary for BIM. To address these issues, this research proposes a hybrid semantic-geometric approach for clutter-resistant floorplan generation from laser-scanned building point clouds. The input point clouds are first pre-processed by normalizing the coordinate system and removing outliers. Then, a semantic segmentation network based on PointNet++ is used to label each point as ceiling, floor, wall, door, stair, and clutter. The clutter points are removed whereas the wall, door, and stair points are used for 2D floorplan generation. A region-growing segmentation algorithm paired with geometric reasoning rules is applied to group the points together into individual building elements. Finally, a 2-fold Random Sample Consensus (RANSAC) algorithm is applied to parameterize the building elements into 2D lines which are used to create the output floorplan. The proposed method is evaluated using the metrics of precision, recall, Intersection-over-Union (IOU), Betti error, and warping error.

• Journal of The Korean Astronomical Society
• /
• v.47 no.6
• /
• pp.319-325
• /
• 2014

We map two molecular clouds located in the exact anticenter region emitting in the (J = 1-0) transition of $^{12}CO$ and $^{13}CO$ using the 3-mm SIS mixer receiver on the 14-m radio telescope at Taeduk Radio Astronomy Observatory. The target clouds with anomalous velocities of $V_{LSR}{\sim}-20km\;s^{-1}$ are distinguished from other clouds in this direction. In addition, they are located in the interarm region between the Orion Arm and the Perseus Arm. Sizes of the clouds are estimated to be about 8.6 and 10.8 pc, respectively. The total mass is estimated to be about $4{\times}10^3$ $M_{\odot}$ using CO luminosity of the clouds. Several cores are detected, but no sign of star formation is found according to the IRAS point sources. Their larger linewidths, anomalous velocities, and their location at the interarm region make these clouds more distinguished, though their physical properties are similar to the dark clouds in the solar neighborhood in terms of mass and size.

• Journal of Computational Design and Engineering
• /
• v.1 no.2
• /
• pp.116-127
• /
• 2014

This paper aims to review methods for computing orthogonal projection of points onto curves and surfaces, which are given in implicit or parametric form or as point clouds. Special emphasis is place on orthogonal projection onto conics along with reviews on orthogonal projection of points onto curves and surfaces in implicit and parametric form. Except for conics, computation methods are classified into two groups based on the core approaches: iterative and subdivision based. An extension of orthogonal projection of points to orthogonal projection of curves onto surfaces is briefly explored. Next, the discussion continues toward orthogonal projection of points onto point clouds, which spawns a different branch of algorithms in the context of orthogonal projection. The paper concludes with comments on guidance for an appropriate choice of methods for various applications.

• Korean Journal of Computational Design and Engineering
• /
• v.19 no.3
• /
• pp.294-304
• /
• 2014

This paper presents a real-time rendering algorithm of large-scale geometric data using GLSL (OpenGL shading language). It details the VAO (vertex array object) and VBO(vertex buffer object) to be used for up-loading the large-scale point clouds and polygon meshes to a graphic video memory, and describes the shader program composed by a vertex shader and a fragment shader, which manipulates those large-scale data to be rendered by GPU. In addition, we explain the global rendering procedure that creates and runs the shader program with the VAO and VBO. Finally, a rendering performance will be measured with application examples, from which it will be demonstrated that the proposed algorithm enables a real-time rendering of large amount of geometric data, almost impossible to carry out by previous techniques.