• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권3호
• /
• pp.1142-1166
• /
• 2020

Salp Swarm Algorithm (SSA) is a new nature-inspired swarm optimization algorithm that mimics the swarming behavior of salps navigating and foraging in the oceans. SSA has been proved to enable to avoid local optima and enhance convergence speed benefiting from the adaptive nonlinear mechanism and salp chains. In this paper, visual tracking is considered to be a process of locating the optimal position through the interaction between leaders and followers in successive images. A novel SSA-based tracking framework is proposed and the analysis and adjustment of parameters are discussed experimentally. Besides, the qualitative analysis and quantitative analysis are performed to demonstrate the tracking effect of our proposed approach by comparing with ten classical tracking algorithms. Extensive comparative experimental results show that our algorithm has good performance in visual tracking, especially for abrupt motion tracking.

• 방송공학회논문지
• /
• 제18권3호
• /
• pp.481-490
• /
• 2013

본 논문에서는 실제 전장에서 전투수행 중인 병사의 개인화기 조준 능력 향상 관점에서 추적기법의 성능평가를 하였다. 실제 전장에서는 짧은 시간동안 전투를 하는 것뿐만 아니라 며칠에 걸쳐서 실시되는 경우도 있다. 이와 같이 장시간 지속되는 작전 중에서 다양한 요소에 의해서 지속적으로 병사의 신체능력이 감소된다. 이렇게 손실되는 신체능력을 보완하기 위하여 시각추적 기술을 화기의 조준경에 적용하여 적 병사 이동상황을 자동적으로 추적하고 이로 인해 감소된 조준능력을 향상시키기 위한 실험을 하였다. 최신영상 추적 기법들 중에서 최적의 것을 결정하기 위하여, 겹침 현상, 카메라 이동, 크기변화, 저대비 영상, 조명변화 등의 특징이 포함된 여러 실제 전장 영상으로 그 성능을 평가하였다. VTD (Visual Tracking Decomposition)[2]가 정확도에서 IVT (Incremental learning for robust Visual Tracking)[7]가 속도 평가에서 가장 우수하였으며 종합적으로는 MIL (Multiple Instance Learning)[1]이 가장 우수한 결과를 보여 주었다. 이러한 성능평가 결과는 시각추적기술이 적용된 조준경이 실제 전장에서 전투수행을 하면서 신체능력이 감소된 병사의 전투력을 보완할 가능성이 있다는 것을 보여 준다.

• Architectural research
• /
• 제22권4호
• /
• pp.155-162
• /
• 2020

Numerous scholars have explored the modeling, control, and optimization of energy systems in buildings, offering new insights about technology and environments that can advance industry innovation. Eye trackers deliver objective eye-gaze data about visual and attentional processes. Due to its flexibility, accuracy, and efficiency in research, eye tracking has a control scheme that makes measuring rapid eye movement in three-dimensional space possible (e.g., virtual reality, augmented reality). Because eye movement is an effective modality for digital interaction with a virtual environment, tracking how users scan a visual field and fix on various digital objects can help designers optimize building environments and materials. Although several scholars have conducted Virtual Reality studies in three-dimensional space, scholars have not agreed on a consistent way to analyze eye tracking data. We conducted eye tracking experiments using objects in three-dimensional space to find an objective way to process quantitative visual data. By applying a 12 × 12 grid framework for eye tracking analysis, we investigated how people gazed at objects in a virtual space wearing a headmounted display. The findings provide an empirical base for a standardized protocol for analyzing eye tracking data in the context of virtual environments.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.798-802
• /
• 2003

The robustness and reliability of vision algorithms is the key issue in robotic research and industrial applications. In this paper, the robust real time visual tracking in complex scene is considered. A common approach to increase robustness of a tracking system is to use different models (CAD model etc.) known a priori. Also fusion of multiple features facilitates robust detection and tracking of objects in scenes of realistic complexity. Because voting is a very simple or no model is needed for fusion, voting-based fusion of cues is applied. The approach for this algorithm is tested in a 3D Cartesian robot which tracks a toy vehicle moving along 3D rail, and the Kalman filter is used to estimate the motion parameters, namely the system state vector of moving object with unknown dynamics. Experimental results show that fusion of cues and motion estimation in a tracking system has a robust performance.

• 한국정밀공학회지
• /
• 제15권7호
• /
• pp.175-181
• /
• 1998

This paper describes realtime visual tracking system of moving object for three dimensional target using EKF(Extended Kalman Filter). We present a new realtime visual tracking using EKF algorithm and image prediction algorithm. We demonstrate the performance of these tracking algorithm through real experiment. The experimental results show the effectiveness of the EKF algorithm and image prediction algorithm for realtime tracking and estimated state value of filter, predicting the position of moving object to minimize an image processing area, and by reducing the effect by quantization noise of image.

• 한국정밀공학회지
• /
• 제22권9호
• /
• pp.77-84
• /
• 2005

In most vision applications, we are frequently confronted with determining the position of object continuously. Generally, intertwined processes ire needed for target tracking, composed with tracking and control process. Each of these processes can be studied independently. In case of actual implementation we must consider the interaction between them to achieve robust performance. In this paper, the robust real time visual tracking in complex background is considered. A common approach to increase robustness of a tracking system is to use known geometric models (CAD model etc.) or to attach the marker. In case an object has arbitrary shape or it is difficult to attach the marker to object, we present a method to track the target easily as we set up the color and shape for a part of object previously. Robust detection can be achieved by integrating voting-based visual cues. Kalman filter is used to estimate the motion of moving object in 3D space, and this algorithm is tested in a pan/tilt robot system. Experimental results show that fusion of cues and motion estimation in a tracking system has a robust performance.

• 제어로봇시스템학회논문지
• /
• 제16권9호
• /
• pp.883-890
• /
• 2010

This paper proposes tip position control system which uses a visual marker to determine the tip position of a robot manipulator. The main idea of this paper is to introduce visual marker for the tracking control of a robot manipulator. Existing researches utilize stationary markers to get pattern information from them. Unlike existing researches, we introduce visual markers to get the coordinates of them in addition to their pattern information. Markers need not be stationary and the extracted coordinate of markers are used as a reference trajectory for the tracking control of a robot manipulator. To build the proposed control scheme, we first obtain intrinsic parameters through camera calibration and evaluate their validity. Secondly, we present a procedure to obtain the relative coordinate of a visual marker with respect to a camera. Thirdly, we derive the equation for the kinematics of the SCORBOTER 4pc manipulator which we use for control of manipulator. Also, we provide a flow diagram of entire visual marker tracking system. The feasibility of the proposed scheme will be demonstrated through real experiments.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2016년도 춘계학술대회
• /
• pp.101-103
• /
• 2016

최근 들어, 영상 추적(Visual Tracking)에서의 목표물을 sparse coefficient vector로 나타낼 수 있게 되면서, L1 minimization 방법을 이용한 영상처리 속도 향상이 필요하게 되었다. 더 나아가서, L1 minimization 방법은 영상 추적 과정에서 주로 발생하는 occlusion 문제를 해결하는 방법으로 많이 사용되고 있다. 다라서 본 논문에서는 영상 추적 과정에서 발생하는 occlusion 문제의 해결을 위해서 L1 minimization의 parameter를 분석하였다. L1 minimization에는 최소화 결과에 영향을 미치는 weighted parameter가 존재하며, 이들은 고정 상수나 목표물의 중간값, 평균값, 표준편차로 나타내어 진다. 실험 결과를 바탕으로 분석하였을 때, weighted parameter 중에서 평균값이 OPE(One Pass Evaluation)을 기반으로 한 success rate와 precision performance에서 좋은 결과를 갖는 것을 확인할 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1995년도 Proceedings of the Korea Automation Control Conference, 10th (KACC); Seoul, Korea; 23-25 Oct. 1995
• /
• pp.131-134
• /
• 1995

We developed a visual system that is able to track the moving objects within a certain range of errors. The visual system is driven by two DC servo motors that are controlled by a computer based on the visual data obtained from a CCD video camera. The software to track the moving objects is developed based on the PWM of the DC motors. Also, the problems how to implement a fuzzy logic control method and a neural network in this system, are also considered in order to check the control performance of tracking. The fuzzy logic algorithm is a powerful control technique for nonlinear dynamical system and also the neural network could be implemented in this system. In this paper, we present configuration of tracking system developed in our laboratory, the control methods of the visual system and the experimental results are shown.

• 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.