• 한국통신학회논문지
• /
• 제28권10C호
• /
• pp.1013-1022
• /
• 2003

본 논문은 연속적인 영상에 대해 비모수적 영상 처리 기법과 칼만 필터 기반의 동적 AR(2) 프로세스 기법을 적용하여 객체의 움직임을 적응적으로 추적하는 실시간 객체 추적 알고리즘을 구현하였다. 다양한 환경 조건에서 입력되는 영상에 대해 비모수적 영상 처리 기법을 이용하여 처리함으로써 효과적으로 움직임 객체를 추출하였으며, 객체의 움직임을 동적 AR(2) 프로세스 모형으로 모델링하고 동적으로 변하는 AR(2) 프로세스의 파라미터를 칼만 필터를 통해 추정함으로써 객체의 다변적인 움직임을 적응적으로 예측하여 추적할 수 있었다. 구현된 객체 추적 시스템을 실험한 결과, 기존의 선형 칼만 필터 기법을 이용한 추적 기법과 비교하여 추정 오차가 약 1/2.5∼1/50 만큼 더 적게 나와 객체의 움직임을 더 근사적으로 추적함을 알 수 있었다.

• 한국공작기계학회논문집
• /
• 제14권4호
• /
• pp.81-88
• /
• 2005

This paper attempts to derive the dynamic model of handling tasks in finger robot which grasps stable and manipulates a rigid object with some dexterity. Firstly, a set of differential equation describing dynamics of the manipulators and object together with geometric constraint of tight area-contacts is formulated by Lagrange's equation. Secondly, the roblems of controlling both the forces of pressing object and the rotation angle of the object under the geometric constraints are discussed. The effect of geometric constraints of area-contacts between the link's end-effector and the object is analyzed and the model based on the differential-algebraic equations is presented. In this paper, the control method for dynamic stable grasping and enhancing dexterity in manipulating things is proposed. It is illustrated by computer simulation and the experiment that the control system gives the performance improvement in the dynamic stable grasping and nimble manipulating of the dual fingers robot with soft tips.

• 대한임베디드공학회논문지
• /
• 제16권6호
• /
• pp.323-330
• /
• 2021

Visual object tracking is a challenging area of study in the field of computer vision due to many difficult problems, including a fast variation of target shape, occlusion, and arbitrary ground truth object designation. In this paper, we focus on the reinforced feature of the dynamic search area to get better performance than conventional discriminative model prediction trackers on the condition when the accuracy deteriorates since low feature discrimination. We propose a reinforced input feature method shown like the spotlight effect on the dynamic search area of the target tracking. This method can be used to improve performances for deep learning based discriminative model prediction tracker, also various types of trackers which are used to infer the center of the target based on the visual object tracking. The proposed method shows the improved tracking performance than the baseline trackers, achieving a relative gain of 38% quantitative improvement from 0.433 to 0.601 F-score at the visual object tracking evaluation.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제26권9호
• /
• pp.1052-1060
• /
• 1999

이제까지 비디오 데이타베이스를 모델링하기 위한 많은 연구들이 수행되었지만 그 모든 모델들에서 다루는 비디오 데이타는 사용자의 개입이 없을 때 항상 미리 정의된 순서로 보여진다는 점에서 정적 데이타 모델로 간주될 수 있다. 주문형 뉴스 서비스, 주문형 비디오 서비스, 디지털 도서관, 인터넷 쇼핑 등과 같이 최신 비디오 정보 서비스를 제공하는 비디오 데이타베이스 응용들에서는 빈번한 비디오 편집이 요구되는데 실시간 처리가 바람직하다. 이를 위해서 기존의 비디오 데이타 내용이 변경되거나 새로운 비디오 데이타가 생성되어야 하지만 이제까지의 비디오 데이타 모델에서는 이러한 비디오 편집 작업이 일일이 수작업으로 수행되어야만 했다. 본 논문에서는 비디오 편집에 드는 노력을 줄이기 위해서 객체지향 데이타 모델에 기반하여 DVID(Dynamic Video Object Data Model)라는 동적 비디오 객체 데이타 모델을 제안한다. DVID는 기존의 정적 비디오 객체뿐만 아니라 사용자의 개입없이도 비디오의 내용을 비디오 데이타베이스로부터 동적으로 결정하여 보여주는 동적 비디오 객체도 함께 제공한다.Abstract A lot of research has been done on modeling video databases, but all of them can be considered as the static video data model from the viewpoint that all video data on those models are always presented according to the predefined sequences if there is no user interaction. For some video database applications which provides with up-to-date video information services such as news-on-demand, video-on-demand, digital library, internet shopping, etc., video editing is requested frequently, preferably in real time. To do this, the contents of the existing video data should be changed or new video data should be created, but on the traditional video data models such video editing works should be done manually. In order to save trouble in video editing work, this paper proposes the dynamic video object data model named DVID based on object oriented data model. DVID allows not only the static video object but also the dynamic video object whose contents are dynamically determined from video databases in real time even without user interaction.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 V
• /
• pp.2705-2708
• /
• 2003

In this paper, we propose a method that applies matrix decomposition technique to the connection of actuator capabilities of each robot to object acceleration limits for multiple cooperative robot systems. The robot systems under consideration are composed of several robot manipulators and each robot contacts a single object to carry the object while satisfying the constraints described in kinematics as well as dynamics. By manipulating kinematic and dynamic equations of both robots and objects, we at first derive a matrix relating joint torques with object acceleration, manipulate the null space of the matrix, and then we decompose the matrix into three parts representing indeterminancy, connectivity, and redundancy. With the decomposed matrix we derive the boundaries of object accelerations from given joint actuators. To show the validity of the proposed method some examples are given in which the results can be expected by intuitive observation.

• 한국자동차공학회논문집
• /
• 제11권5호
• /
• pp.147-155
• /
• 2003

The application of object-oriented modeling to develop a powertrain performance simulation program, called O-DYN, is introduced. Powertrain components, such as the engine, transmission, shaft, clutch are modeled as classes which have data and method by using object-oriented modeling methodology. O-DYN, a performance simulation program, based on the object-oriented modeling is made in C++. One powertrain simulation using O-DYN is proposed, and it is expected that the simulation program or individual class constructed in this paper would be useful for automotive engineers to predict the performance of powertrains and to develop a simulation program.

• International Journal of Control, Automation, and Systems
• /
• 제2권2호
• /
• pp.228-237
• /
• 2004

This paper provides a motion analysis of soft-fingertip object manipulation tasks by presenting a dynamic model of multi-fingered object manipulations with soft fingertips. It is fundamentally observed that soft fingertips employed in a multi-fingered hand generate some deformation effects during the manipulation process and also that those effects are closely related to the behavior of the manipulated object. In order to analyze the motion of using soft fingertips, a dynamic manipulation control scheme is presented. Simulation and experimental results demonstrate the motion of soft-fingertips applied in object manipulating tasks and are further used to discuss the characteristics of soft-fingertip motions.

• 제어로봇시스템학회논문지
• /
• 제12권8호
• /
• pp.746-751
• /
• 2006

In this paper, we present tracking algorithm for the indoor moving object. We research passive method using a camera and image processing. It had been researched to use dynamic based estimators, such as Kalman Filter, Extended Kalman Filter and Particle Filter for tracking moving object. These algorithm have a good performance on real-time tracking, but they have a limit. If the shape of object is changed or object is located on complex background, they will fail to track them. This problem will need the complicated image processing algorithm. Finally, a large algorithm is made from integration of dynamic based estimator and image processing algorithm. For eliminating this inefficiency problem, image based estimator, Mean-shift Algorithm is suggested. This algorithm is implemented by color histogram. In other words, it decide coordinate of object's center from using probability density of histogram in image. Although shape is changed, this is not disturbed by complex background and can track object. This paper shows the results in real camera system, and decides 3D coordinate using the data from mean-shift algorithm and relationship of real frame and camera frame.

• 한국지능시스템학회논문지
• /
• 제15권5호
• /
• pp.626-631
• /
• 2005

본 논문에서는 Dynamic Programming(DP)을 적용한 트리구조 미로내의 목표물 탐색 알고리즘을 구현한다. DP는 큰 문제를 이루는 작은 문제들을 먼저 해결하고 작은 문제들의 최적해를 이용하여 순환적으로 큰 문제를 해결한다. 먼저 실험을 위해 적외선 센서론 부착한 소형 이동 로봇과, 'Y'형태로 갈라진 길을 연결한 트리 구조의 미로 환경을 구성한다. 실험에서는 두 개의 서로 다른 알고리즘 - 좌수법, DP - 을 사용하여 목표물 탐색을 시도한다. 마지막으로 위 실험을 통해 DP를 미로 탐색문제에 적용했을 때의 성능을 검증한다.

• 한국멀티미디어학회논문지
• /
• 제21권5호
• /
• pp.558-570
• /
• 2018

Recently, object detection is a critical function for any system that uses computer vision and is widely used in various fields such as video surveillance and self-driving cars. However, the conventional methods can not detect the objects clearly because of the dynamic background change in the beach. In this paper, we propose a new technique to detect humans correctly in the dynamic videos like shores. A new background modeling method that combines spatial GMM (Gaussian Mixture Model) and temporal GMM is proposed to make more correct background image. Also, the proposed method improve the accuracy of people detection by using SVM (Support Vector Machine) to classify people from the objects and KCF (Kernelized Correlation Filter) Tracker to track people continuously in the complicated environment. The experimental result shows that our method can work well for detection and tracking of objects in videos containing dynamic factors and situations.