• 정보처리학회 논문지
• /
• 제13권7호
• /
• pp.335-347
• /
• 2024

개방형 어휘 3차원 포인트 클라우드 개체 분할은 3차원 장면 포인트 클라우드를 훈련단계에서 등장하였던 기본 클래스의 개체들뿐만 아니라 새로운 신규 클래스의 개체들로도 분할해야 하는 어려운 시각적 작업이다. 본 논문에서는 중요한 모델 설계 이슈별 기존 모델들의 한계점들을 극복하기 위해, 새로운 개방형 어휘 3차원 개체 분할 모델인 Open3DME를 제안한다. 첫째, 제안 모델은 클래스-독립적인 3차원 마스크의 품질을 향상시키기 위해, 새로운 트랜스포머 기반 3차원 포인트 클라우드 개체 분할 모델인 T3DIS[6]를 마스크 제안 모듈로 채용한다. 둘째, 제안 모델은 각 포인트 세그먼트별로 텍스트와 의미적으로 정렬된 시각적 특징을 얻기 위해, 사전 학습된 OpenScene 인코더와 CLIP 인코더를 적용하여 포인트 클라우드와 멀티-뷰 RGB 영상들로부터 각각 3차원 및 2차원 특징들을 추출한다. 마지막으로, 제안 모델은 개방형 어휘 레이블 할당 과정동안 각 포인트 클라우드 세그먼트별로 추출한 2차원 시각적 특징과 3차원 시각적 특징을 상호 보완적으로 함께 이용하기 위해, 특징 앙상블 기법을 적용한다. 본 논문에서는 ScanNet-V2 벤치마크 데이터 집합을 이용한 다양한 정량적, 정성적 실험들을 통해, 제안 모델의 성능 우수성을 입증한다.

• 한국군사과학기술학회지
• /
• 제23권6호
• /
• pp.542-553
• /
• 2020

Virtual world and simulation need large scale map rendering. However, rendering too many vertices is a computationally complex and time-consuming process. Some game development companies have developed 3D LOD objects for high-speed rendering based on distance between camera and 3D object. Terrain physics simulation researchers need a way to recognize the original object shape from 3D LOD objects. In this paper, we proposed simply automatic LOD framework using point cloud data (PCD). This PCD was created using a 6-direct orthographic ray. Various experiments are performed to validate the effectiveness of the proposed method. We hope the proposed automatic LOD generation framework can play an important role in game development and terrain physic simulation.

• 대한화학회지
• /
• 제59권5호
• /
• pp.407-412
• /
• 2015

A preconcentration procedure for determination of palladium by laser thermal lens spectrometry (TLS) is proposed. It is based on cloud point extraction of palladium(II) ions as 2-(3,5-dichloro-2-pyridylazo)-5-dimethylaminoaniline (3,5-diCl-PADMA) complexes using octylphenoxypolyethoxyethanol (Triton X-114) as surfactant. The effects of various experimental conditions such as pH, concentration of ligand and surfactant, equilibration temperature and time on cloud point extraction were studied. Under the optimized conditions, the calibration graph was linear in the range of 0.15~6 ng mL⁻¹, and the detection limit was 0.04 ng mL⁻¹ with an enrichment factor of 22. The sensitivity was enhanced by 846 times when compared with the conventional spectrophotometric method. The recovery of palladium was in the range of 96.6%~104.0%. The proposed method was applied to the determination of palladium in water samples.

• 전자공학회논문지
• /
• 제52권12호
• /
• pp.77-86
• /
• 2015

LIDAR를 이용해서 얻은 3차원 점군 데이터는 작은 물체를 추출하기에는 오차의 영향이 크기 때문에 작은 밸브를 자동으로 추출하는데 많은 어려움이 있다. 본 논문에서는 이러한 잡음이 있는 3차원 점군 데이터 사이에서 밸브의 위치 및 방향(Pose)의 정보를 얻는 방법을 제안한다. Pose를 얻기 위해서 밸브가 원환체 모양의 손잡이, 원통 모양의 Rib, 평면 모양의 중심축 평면인 기본 도형으로 이루어진 모델이라고 가정한다. 그리고 밸브의 중심 좌표에 대한 추가적인 입력을 받아서 밸브의 Pose를 추출한다. 중심점을 기준으로 거리에 따른 히스토그램을 생성하고, 히스토그램의 값에 따라 손잡이, Rib, 중심축 평면의 파라미터를 통계적인 방법으로 추출하여 최종 밸브의 Pose를 추출한다. 추출된 밸브의 Pose를 이용하여 3차원 점군 데이터에 밸브의 모형을 각 모양으로 복원한다.

• 한국CDE학회논문집
• /
• 제21권3호
• /
• pp.341-352
• /
• 2016

Manual 3D pipeline modeling from LiDAR scanned point cloud data is laborious and time-consuming process. This paper presents a method to extract the pipe, elbow and branch information which is essential to the automatic modeling of the pipeline connection. The pipe geometry is estimated from the point cloud data through the Hough transform and the elbow position is calculated by the medial axis intersection for assembling the nearest pair of pipes. The branch is also created for a pair of pipe segments by estimating the virtual points on one pipe segment and checking for any feasible intersection with the other pipe's endpoint within the pre-defined range of distance. As a result of the automatic modeling, a complete 3D pipeline model is generated by connecting the extracted information of pipes, elbows and branches.

• Macromolecular Research
• /
• 제11권4호
• /
• pp.231-235
• /
• 2003

Phase behavior information is necessary for accomplishing homogeneous copolymerization to obtain high yield of copolymers and prevent a fouling problem. Cloud-point data to $160^{\circ}C$ and 1,450 bar are presented for five $C_6$ hydrocarbon solvents, normal hexane, 2,2-dimethyl butane, 2,3-dimethyl butane, 2-methyl pentane, and 3-methyl pentane, with poly(ethylene-co-53 mol% norbornene) ($PEN_{53}$). The pressure-concentration isotherms measured for $PEN_{53}$/n-hexane have maximums that range between 5 and 12 wt% $PEN_{53}$. The cloud-point curves for $PEN_{53}$ all have negative slopes that decrease in pressure with temperatures. The single-phase region of $PEN_{53}$ in n-hexane is larger than the regions in 2,2-dimethyl butane, 2,3-dimethyl butane, 2-methyl pentane, and 3-methyl pentane. The cloud-point curve of $PEN_{53}$ in 2,2-dimethyl butane is located at higher temperatures and pressures than the curve in 2,3-dimethyl butane due to the reduced dispersion interactions with and limited access of 2,2-dimethyl butane to the copolymer. Similar cloud-point behavior is observed for $PEN_{53}$ in 2-methyl pentane and 3-methyl pentane.

• Bulletin of the Korean Chemical Society
• /
• 제25권8호
• /
• pp.1133-1136
• /
• 2004

The need for highly reliable methods for the determination of trace metals is recognized in analytical chemistry and environmental science. A method based on the cloud-point extraction (CPE) technique for the trace analysis of Pb and Cu in water samples is described in this study. The analytes in the initial aqueous solution are complexed with pyrogallol, and 0.1%(w/v) Triton X-114 is added as surfactant. Following phase separation at $50^{\circ}C$, based on the cloud point of the mixture and dilution of the surfactant-rich phase with acidified methanolic solution, the enriched analytes are determined by flame atomic absorption spectrometry. After optimization of the complexation and extraction conditions, the enrichment factors of Pb and Cu were found to be 72 and 85, respectively. Under optimum conditions, the preconcentration of 60 mL of samples in the presence of 0.1%(w/v) Triton X-114 permitted the detection of 0.4 ${\mu}gL^{?1}$ of Pb and 0.05 ${\mu}gL^{?1}$ of Cu. The proposed method was applied successfully to the determination of Pb and Cu in water samples.

• 대한조선학회논문집
• /
• 제59권4호
• /
• pp.235-242
• /
• 2022

As ship condition inspection technology has been developed, research on collecting, analyzing, and diagnosing condition information has become active. In ships, related research has been conducted, such as analyzing, detecting, and classifying major hull failures such as cracks and corrosion using 2D and 3D data information. However, for geometric deformation such as indents and bulges, 2D data has limitations in detection, so 3D data is needed to utilize spatial feature information. In this study, we aim to detect hull structural deformation positions. It builds a specimen based on actual hull structure deformation and acquires a point cloud from a model scanned with a 3D scanner. In the obtained point cloud, deformation(outliers) is found with a combination of RANSAC algorithms that find the best matching model in the Octree data structure and dataset.

• 한국BIM학회 논문집
• /
• 제14권2호
• /
• pp.1-12
• /
• 2024

The paper proposes a methodology that combines the Random Sample Consensus (RANSAC) algorithm and the Point Cloud Encoder-Decoder Network (PCEDNet) algorithm to automatically extract the length of infrastructure elements from point cloud data acquired through 3D LiDAR scans of retaining walls. This methodology is expected to significantly improve time and cost efficiency compared to traditional manual measurement techniques, which are crucial for the data-driven analysis required in the precision-demanding construction sector. Additionally, the extracted positional and dimensional data can contribute to enhanced accuracy and reliability in Scan-to-BIM processes. The results of this study are anticipated to provide important insights that could accelerate the digital transformation of the construction industry. This paper provides empirical data on how the integration of digital technologies can enhance efficiency and accuracy in the construction industry, and offers directions for future research and application.

• 한국측량학회지
• /
• 제31권1호
• /
• pp.41-48
• /
• 2013

본 연구는 대용량 3차원 포인트 클라우드의 탐색을 위한 메모리 효율적인 옥트리를 설계하는 것을 목표로 한다. 이를 위하여 C++ 언어로 구현된 옥트리 노드의 구성요소 중 최소 경계 입방체 좌표 정보 등을 위한 변수를 제거하는 대신, 부모 노드에서 자식 노드 접근시 최소 경계 입방체 좌표를 계산하여 전달하였다. 아울러 자식 노드 등의 생성시마다 new 연산자를 사용하는 대신, 수도 트리와 정식 트리를 생성하는 이중적인 과정을 통해 노드 등을 배열로 미리 선언함으로서 메모리 효율성을 더욱 개선하였다. 1800만개 이상의 포인트로 구성된 실제 포인트 클라우드를 대상으로 트리를 구성하고 인접 포인트를 탐색하는 실험을 수행하였다. 최소 경계 입방체 좌표 정보를 노드에 저장하는 경우와 비교한 결과 메모리 사용량과 탐색 속도의 트레이드오프가 존재하지만 제안한 방식이 비교군보다 메모리 효율적이어서 대용량 포인트 클라우드에 적용할 수 있는 대안임을 확인할 수 있었다.