• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.119-122
• /
• 2016

최근 IT기술 영역에서 미래기술로 촉망받는 증강현실(AR)과 가상현실(VR)환경을 구축함에 있어서, 마우스나 키보드 등의 별도 장치 없이 기기에 원하는 동작을 입력 하도록 하는 NUI(Natural User Interface)기술이 각광받고 있다. 또한 NUI를 구현하는데 중요한 기술 중 하나로 손동작 인식 기술, 얼굴 인식 기술 등이 대두되고 있다. 이에 본 논문은 적외선 센서의 일종인 Leapmotion 센서를 사용하여 손동작 인식을 구현하고자 하였다. 첫 번째로 우선 거리변환 행렬을 사용하여 손바닥의 중심을 찾았다. 이후 각각의 손가락을 convex hull 알고리즘을 사용하여 추출한다. 제안한 알고리즘에서는 손가락, 손바닥 부분의optical flow를 구한 후, 두 optical flow의 특성을 사용하여 손의 이동, 정지, 클릭 동작을 구분 할 수 있도록 하였다.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.18 no.6
• /
• pp.576-584
• /
• 2009

Up to now, it has been said that no satisfactory computer solution has been found for synthesizing four-bar linkage based on the prescribed coupler link curve. In our study, an algorithm has been developed to improve the design synthesis of four bar linkage based on the 5 precision points method. The suggested algorithm generates the desired coupler curve by using NURBS, and then the generated curve approximates as closely as possible to the desired curve representing coupler link trajectory. Also, when comparing each generated curve by constructing the control polygon, rapid comparison is easily achieved by applying convex hull of the control polygon. Finally, an optimization process using ADS is incorporated into the algorithm based on the 5 precision point method to reduce the total optimization process time. As for examples, two four bar linkages were tested and the result well demonstrated the effectiveness of the algorithm.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.323-326
• /
• 2014

본 논문에서는 깊이 카메라를 이용한 사용자의 손 모양, 특히 수화를 인식하는 방법에 대해 제안한다. 손 모양 인식은 손가락 검출과 손 인식으로 크게 2가지로 나눌 수 있다. 손가락 검출을 위해 본 시스템에서는 Distance Transform을 이용하여 손의 뼈대를 검출 하고, Convex Hull을 통해 손가락을 검출하는 방법을 제안한다. 뼈대 검출은 보다 정확한 손가락을 검출할 수 있는 장점이 생긴다. 손 인식에는 손 중심과 손가락의 길이, 손의 축, 손가락의 축, 팔 중심의 위치 등을 이용하여 Decision Tree를 생성하고, 반복적 검사를 통해 인식의 오류율을 줄였다. 실험결과에서는 수화 인식이 성공적으로 잘 인식 되었다는 것을 보인다.

• Journal of IKEEE
• /
• v.19 no.4
• /
• pp.526-534
• /
• 2015

In this paper, we studied the use of the foreground and background features and SVM classifier to improve the accuracy of offline handwritten numeral recognition. The foreground features are two directional features: directional gradient feature by Kirsch operators and directional stroke feature by local shrinking and expanding operations, and the background feature is concavity feature which is extracted from the convex hull of the numeral, where the concavity feature functions as complement to the directional features. During classification of the numeral, these three features are combined to obtain good discrimination power. The efficiency of our scheme is tested by recognition experiments on the handwritten numeral database CENPARMI, where SVM classifier with RBF kernel is used. The experimental results show the usefulness of our scheme and recognition rate of 99.10% is achieved.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.440-442
• /
• 2000

필기 한글 문자 인식을 위해서는 패턴을 구성하는 획 성분을 분석하는 작업이 필수적이다. 획 성분 추출을 위해 사용한 세선화 방법은 입력 영상을 왜곡하는 단점을 가지고 있다. 이를 극복하기 위하여 본 논문은 입력 영상을 왜곡하지 않고 의미 있는 부품 단위로 분할하는 방법을 제안한다. 의미 있는 부품이란 유사 볼록하게 분할된 영역을 의미한다. 분할 방법은 먼저 입력 영상에 볼록 헐 연산을 적용하여 오목 영역을 생성한다. 이 오목 영역에서 분할 기준(anchor point)점을 탐지하고 획의 반대편 외곽선 상에서 분할 끝(terminal point)점을 찾아 분할 경로를 구성하여 획을 분할한다. 모든 부품이 유사 볼록 조건을 만족할 때까지 위 과정을 반복 수행한다. 제안한 방법은 두 개의 파라미터만을 가지며 간단한 프로시져로 구성되어 있다. 또한 필기 한글 패턴뿐 아니라 여러 언어에 적용 가능하다는 장점을 갖는다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.18 no.2
• /
• pp.43-49
• /
• 1992

Edge coloring problem is to find a minimum cardinality coloring of the edges of a graph so that any pair of edges incident to a common node do not have the same colors. Edge coloring problem is NP-hard, hence it is unlikely that there exists a polynomial time algorithm. We formulate the problem as a covering of the edges by matchings and find valid inequalities for the convex hull of feasible solutions. We show that adding the valid inequalities to the linear programming relaxation is enough to determine the minimum coloring number(chromatic index). We also propose a method to use the valid inequalities as cutting planes and do the branch and bound search implicitly. An example is given to show how the method works.

• Journal of Mechanical Science and Technology
• /
• v.18 no.7
• /
• pp.1236-1245
• /
• 2004

In this study, an optical imaging method was developed for the measurements of the sizes and velocities of droplets in sprays. Double-exposure single-frame spray images were captured by the imaging system. An image processing program was developed for the measurements of the sizes and positions of individual particles including separation of the overlapped particles and particle tracking and pairing at two time instants. To recognize and separate overlapping particles, the morphological method based on watershed segmentation as well as separation using the perimeter and convex hull of image was used consecutively. Better results in separation were obtained by utilization of both methods especially for the multiple or heavily-overlapped particles. The match probability method was adopted for particle tracking and pairing after identifying the positions of individual particles and it produced good matching results even for large particles like droplets in sprays. Therefore, the developed optical imaging method could provide a reliable way of analyzing the motion and size distribution of droplets produced by various sprays and atomization devices.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2015.10a
• /
• pp.1776-1777
• /
• 2015

본 연구에서는 3D 객체의 수직 방향(upright orientation)과 오버행(overhang) 면적을 고려하는 3D 프린팅 방향 최적화 방법을 제안한다. 모든 메시 법선들을 세 그룹으로 분류함으로써 객체의 수직방향을 계산한 후, 객체의 무게 중심과 3D convex hull을 사용하여 세 개의 후보 방향을 결정한다. 각 후보 방향에 대하여 오버행 메시 면적을 계산하고 최소 면적을 갖는 후보를 최종 프린팅 방향으로 결정한다. 후보 방향들을 적용하여 회전시킨 모델을 Cura에서 로드하여 프린팅 시간을 가측정한 결과, 제안 방법에 의해 최적화된 방향이 가장 짧은 시간이 소모되는 것으로 나타났다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.3
• /
• pp.163-169
• /
• 2008

In this paper, we propose a real-time people tracking system with multiple CCD cameras for security inside the building. The camera is mounted from the ceiling of the laboratory so that the image data of the passing people are fully overlapped. The implemented system recognizes people movement along various directions. To track people even when their images are partially overlapped, the proposed system estimates and tracks a bounding box enclosing each person in the tracking region. The approximated convex hull of each individual in the tracking area is obtained to provide more accurate tracking information. To achieve this goal, we propose a method for 3D walking human tracking based on the IMPRESARIO framework incorporating cascaded classifiers into hypothesis evaluation. The efficiency of adaptive selection of cascaded classifiers have been also presented. We have shown the improvement of reliability for likelihood calculation by using cascaded classifiers. Experimental results show that the proposed method can smoothly and effectively detect and track walking humans through environments such as dense forests.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.359-364
• /
• 2022

This study presents a nonlinear model predictive control (NMPC)-based obstacle avoidance and whole-body motion planning method for the mobile manipulators. For the whole-body motion control, the mobile manipulator with an omnidirectional mobile base was modeled as a nine degrees-of-freedom (DoFs) serial open chain with the PPR (base) plus 6R (arm) joints, and a swept sphere volume (SSV) was applied to define a convex hull for collision avoidance. The proposed receding horizon control scheme can generate a trajectory to track the end-effector pose while avoiding the self-collision and obstacle in the task space. The proposed method could be calculated using an interior-point (IP) method solver with 100[ms] sampling time and ten samples of horizon size, and the validation of the method was conducted in the environment of Pybullet simulation.