• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2016년도 추계학술발표대회
• /
• pp.733-734
• /
• 2016

This study proposes a novel approach for ground segmentation of 3D point cloud. We combine two techniques: gradient threshold segmentation, and mean height evaluation. Acquired 3D point cloud is represented as a graph data structures by exploiting the structure of 2D reference image. The ground parts nearing the position of the sensor are segmented based on gradient threshold technique. For sparse regions, we separate the ground and nonground by using a technique called mean height evaluation. The main contribution of this study is a new ground segmentation algorithm which works well with 3D point clouds from various environments. The processing time is acceptable and it allows the algorithm running in real time.

• 산업공학
• /
• 제9권1호
• /
• pp.13-24
• /
• 1996

Quality of injection molded parts is dependent on both mold design and processing conditions. From the mold design point of view, an optimal shrinkage should be used to compensate the shrinkage of molded parts. From the processing point of view, it is important to analyze the priority of processing conditions because a number processing conditions affect the quality of molded parts. Processing analysis employing the design of experiment was performed, and the shrinkage of molded part was considered as a characteristic parameter to improve the quality. As the result of the analysis of variance on SN ratio of a characteristic value, injection speed and bolding pressure were selected as two effective process parameters. Regression analysis on shrinkage and processing conditions was carried out, and an optimal processing condition was obtained by the response surface analysis. Shrinkage at the optimal condition could be used to reduce the number of try-cut at the step of mold making. The ranges of indirect control parameter, such as maximum cavity pressure or weight, measured at the optimal processing condition were used for monitoring the quality of molded parts in process.

• 대한임베디드공학회논문지
• /
• 제10권1호
• /
• pp.49-55
• /
• 2015

In this paper, we propose a fast interest point detection method using SURF algorithm. Since the SURF algorithm needs a great computations to detect the interest points and obtain the corresponding descriptors, it is not suitable for real-time based applications. In order to overcome this problem, the interest point detection step is parallelized by OpenMP and SIMD based on analysis of the scale space representation process and localization one in the step. The simulation results demonstrate that processing speed is enhanced about 55% by applying the proposed method.

• 제어로봇시스템학회논문지
• /
• 제9권9호
• /
• pp.714-719
• /
• 2003

We propose a measuring method of large object using the pattern matching. It is hard and expensive to get the complete 3D data when the object is large and exceeds the limit of measuring devices. The large object is divided into several smaller areas and is scanned several times to get the data of all the pieces. These data are aligned to get the complete 3D data using the pattern matching method such as point pattern matching method and transform matrix algorithm. The laser slit beam and CCD camera are applied for the experimental measurement. Visual C＋＋ on Windows 98 is implemented in processing the algorithm.

• 한국정보처리학회논문지
• /
• 제1권3호
• /
• pp.409-417
• /
• 1994

고도행렬로부터 등고선을 추출하는데 있어, 가장 중요한 문제는 하나의 격자에 네개의 통과점이 발생하는 이상 상황(degenerate case)의 해결이다. 본 논문은 이상 상황 검사와 다음점 결정을 동시에 수행하는 방법을 기술한다. 이 방법은 최소 횟수 의 배열 인덱싱을 필요로 한다. 또한 통과점을 모서리점의 높이차에 따라 LOWER/ HIGHER로 표시하고, 이 정보를 이용하여 보다 적은 횟수의 배열 인덱싱을 필요로하는 방법을 제안한다. 그리고 통과점을 표시하고 이들을 추적하기에 적합한 자료구조를 구체적으로 기술한다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2002년도 춘계학술발표논문집 (하)
• /
• pp.1033-1036
• /
• 2002

본 연구에서는 Newton 기법과 모멘트에 기초를 둔 fixed point 알고리즘의 신경망 기반 독립성분분석기법을 제안하였다. 여기서 Newton 기법은 함수의 접선에 기초를 둔 해를 구하는 방법으로 역혼합행렬의 빠른 경신을 위함이고, 모멘트는 접선을 구하는 과정에서 함수의 기울기변화 계산으로 발생하는 발진을 줄여 좀 더 빠른 학습을 위함이다. 제안된 기법을 $256{\times}256$ 픽셀(pixel)의 12개 지문영상으로부터 임의의 혼합행렬에 따라 발생되는 영상들을 각각 대상으로 시뮬레이션 한 결과, 기존의 Fixed point 알고리즘에 의한 결과보다 우수한 분리성능과 빠른 학습속도가 있음을 확인하였다.

• 전기학회논문지
• /
• 제56권8호
• /
• pp.1511-1513
• /
• 2007

A new reference point location method using the cosine component is proposed, where an edge map is defined and used to find the reference point. Because all processes used in the proposed method are performed at the block level, less processing time is required. Experimental results show that the proposed method can effectively detect the reference point with higher speed and accuracy for all types of fingerprints.

• Journal of Information Processing Systems
• /
• 제16권1호
• /
• pp.120-131
• /
• 2020

In order to solve the problem of point clouds coordinate conversion of non-directional scanners, this paper proposes a basic Rodrigues rotation method. Specifically, we convert the 6 degree-of-freedom (6-DOF) rotation and translation matrix into the uniaxial rotation matrix, and establish the equation of objective vector conversion based on the basic Rodrigues rotation scheme. We demonstrate the applicability of the new method by using a bar-shaped emboss point clouds as experimental input, the three-axis error and three-term error as validate indicators. The results suggest that the new method does not need linearization and is suitable for optional rotation angle. Meanwhile, the new method achieves the seamless splicing of point clouds. Furthermore, the coordinate conversion scheme proposed in this paper performs superiority by comparing with the iterative closest point (ICP) conversion method. Therefore, the basic Rodrigues rotation method is not only regarded as a suitable tool to achieve the conversion of point clouds, but also provides certain reference and guidance for similar projects.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2020년도 봄 학술논문 발표대회
• /
• pp.40-41
• /
• 2020

As the capacity of the 3d scanner developed, the reverse engineering using the 3d scanner is emphasized in the construction industry to obtain the 3d geometric representation of buildings. However, big size of the indoor point cloud data acquired by the 3d scanner restricts the efficient process in the reverse engineering. In order to solve this inefficiency, several pre-processing methods simplifying and denoising the raw point cloud data by the rough standard are developed, but these non-standard methods can cause the inaccurate recognition and removal the key-points. This paper analyzes the correlation between the accuracy of wall recognition and the density of the data, thus proposes the proper method for the raw point cloud data. The result of this study could improve the efficiency of the data processing phase in the reverse engineering for indoor point cloud data.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2020년도 춘계학술발표대회
• /
• pp.532-533
• /
• 2020

원격 및 가상환경에서 사용자의 동작에 따른 3D 모델을 제공하기 위해 light detection and range (LiDAR)로 측정된 3D point cloud로 사용자의 3D 모델이 생성되어 원격 및 가상환경에 사용자의 모습이 제공된다. 하지만 3D 모델을 생성하기 위해서는 사용자의 신체 전부가 측정된 3D point cloud가 필요하다. 사용자의 신체 전체를 측정하기 위해서는 적어도 두 개 이상의 LiDAR가 필요하다. 두 개 이상의 LiDAR을 사용할 경우에는 LiDAR을 사용할 공간과 LiDAR를 구비하기 위한 비용이 발생한다. 단일 LiDAR로 3D 모델을 생성하는 방법이 요구된다. 본 논문에서는 단일 LiDAR에서 측정된 3D point cloud를 이용하여 3D 모델을 생성하는 방법이 제안된다. End-to-End 기반 Convolutional Neural Network (CNN) 모델로 측정된 3D point cloud를 분석하여 사용자의 체형과 자세를 예측하도록 학습한다. 기본자세를 취하는 동안 수집된 3D point cloud로 기본이 되는 사용자의 3D 모델을 생성한다. 학습된 CNN 모델을 통하여 측정된 3D point cloud로 사용자의 자세를 예측하여 기본이 되는 3D 모델을 수정하여 3D 모델을 제공한다.