• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.111.4-111
• /
• 2001

In this paper, an adaptively tuned dynamic absorber is proposed. The adaptively tuned dynamic absorber is a dynamic absorber whose stiffness is tuned so that the natural frequency of the absorber coincides with the operating or natural frequency estimated by an adaptive algorithm. The feature of this absorber is as follows. It has an electrodynamic device for the stiffness control. Using Lorenz´s force, it changes the stiffness by changing the applied current. The change of stiffness results in the natural frequency shift, because its mass and damping coefficient are fixed. We may reduce the vibration of the overall system by tuning the natural frequency of the dynamic absorber to the resonant frequency of the structure, when the dominant single tone oscilation occurs in the system ...

• 로봇학회논문지
• /
• 제17권3호
• /
• pp.334-338
• /
• 2022

In visual simultaneous localization and mapping (vSLAM), the correct recognition of a place benefits in relocalization and improved map accuracy. However, its performance is significantly affected by the environmental conditions such as variation in light, viewpoints, seasons, and presence of dynamic objects. This research addresses the problem of feature occlusion caused by interference of dynamic objects leading to the poor performance of visual place recognition algorithm. To overcome the aforementioned problem, this research analyzes the role of scene semantics in correct detection of a place in challenging environments and presents a semantics aided visual place recognition method. Semantics being invariant to viewpoint changes and dynamic environment can improve the overall performance of the place matching method. The proposed method is evaluated on the two benchmark datasets with dynamic environment and seasonal changes. Experimental results show the improved performance of the visual place recognition method for vSLAM.

• Journal of Information Processing Systems
• /
• 제19권1호
• /
• pp.80-88
• /
• 2023

RGB-thermal (RGB-T) tracking using unmanned aerial vehicles (UAVs) involves challenges with regards to the similarity of objects, occlusion, fast motion, and motion blur, among other issues. In this study, we propose dynamic tracking aggregation (DTA) as a unified framework to perform object detection and data association. The proposed approach obtains fused features based a transformer model and an L1-norm strategy. To link the current frame with recent information, a dynamically updated embedding called dynamic tracking identification (DTID) is used to model the iterative tracking process. For object association, we designed a long short-term tracking aggregation module for dynamic feature propagation to match spatial and temporal embeddings. DTA achieved a highly competitive performance in an experimental evaluation on public benchmark datasets.

• Journal of the korean society of oceanography
• /
• 제39권1호
• /
• pp.72-95
• /
• 2004

The Yellow Sea is characterized by relatively shallow water depth, varying range of tidal action and very complex coastal geometry such as islands, bays, peninsulas, tidal flats, shoals etc. The dynamic system is controlled by tides, regional winds, river discharge, and interaction with the Kuroshio. The circulation, water mass properties and their variability in the Yellow Sea are very complicated and still far from clear understanding. In this study, an effort to improve our understanding the dynamic feature of the Yellow Sea system was conducted using numerical simulation with the ROMS model, applying climatologic forcing such as winds, heat flux and fresh water precipitation. The inter-annual variability of general circulation and thermohaline structure throughout the year has been obtained, which has been compared with observational data sets. The simulated horizontal distribution and vertical cross-sectional structures of temperature and salinity show a good agreement with the observational data indicating significantly the water masses such as Yellow Sea Warm Water, Yellow Sea Bottom Cold Water, Changjiang River Diluted Water and other sporadically observed coastal waters around the Yellow Sea. The tidal effects on circulation and dynamic features such as coastal tidal fronts and coastal mixing are predominant in the Yellow Sea. Hence the tidal effects on those dynamic features are dealt in the accompanying paper (Kim et at., 2004). The ROMS model adopts curvilinear grid with horizontal resolution of 35 km and 20 vertical grid spacing confirming to relatively realistic bottom topography. The model was initialized with the LEVITUS climatologic data and forced by the monthly mean air-sea fluxes of momentum, heat and fresh water derived from COADS. On the open boundaries, climatological temperature and salinity are nudged every 20 days for data assimilation to stabilize the modeling implementation. This study demonstrates a Yellow Sea version of Atlantic Basin experiment conducted by Haidvogel et al. (2000) experiment that the ROMS simulates the dynamic variability of temperature, salinity, and velocity fields in the ocean. However the present study has been improved to deal with the large river system, open boundary nudging process and further with combination of the tidal forcing that is a significant feature in the Yellow Sea.

• 한국전산구조공학회논문집
• /
• 제29권4호
• /
• pp.301-308
• /
• 2016

본 연구에서는 마커없이 구조물의 변위를 측정할 수 있는 영상기반 변위계측 시스템(NVDMS)을 제안한다. 기존의 방식과 제안하는 NVDMS는 크게 두 가지의 차이점이 있다. 첫째, NVDMS는 마커를 사용하지 않고 구조물의 특징점의 픽셀좌표 변화를 추출한다. 둘째, 특징점의 픽셀좌표를 물리좌표로 변환하는 scaling factor는 기존 방식에선 마커의 크기로부터 계산되는 반면, NVDMS에서는 카메라와 구조물사이의 거리, 각도, 초점거리로 계산된다. 3층 축소모형의 자유진동 실험에서 제안한 NVDMS로부터 얻은 변위데이터의 신뢰도를 분석하기 위해 LDS로부터 얻은 변위데이터의 비교를 하였으며, 얻어진 변위데이터를 이용하여 동특성을 분석하였다. 분석결과 NVDMS는 마커없이 구조물의 동적변위를 정밀하게 측정가능할 뿐만 아니라 얻어진 변위데이터로부터 추출한 동특성의 신뢰도 또한 높았다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제2권2호
• /
• pp.120-133
• /
• 2008

This paper presents a novel approach for facial motion tracking and facial expression cloning to create a realistic facial animation of a 3D avatar. The exact head pose estimation and facial expression tracking are critical issues that must be solved when developing vision-based computer animation. In this paper, we deal with these two problems. The proposed approach consists of two phases: dynamic head pose estimation and facial expression cloning. The dynamic head pose estimation can robustly estimate a 3D head pose from input video images. Given an initial reference template of a face image and the corresponding 3D head pose, the full head motion is recovered by projecting a cylindrical head model onto the face image. It is possible to recover the head pose regardless of light variations and self-occlusion by updating the template dynamically. In the phase of synthesizing the facial expression, the variations of the major facial feature points of the face images are tracked by using optical flow and the variations are retargeted to the 3D face model. At the same time, we exploit the RBF (Radial Basis Function) to deform the local area of the face model around the major feature points. Consequently, facial expression synthesis is done by directly tracking the variations of the major feature points and indirectly estimating the variations of the regional feature points. From the experiments, we can prove that the proposed vision-based facial expression cloning method automatically estimates the 3D head pose and produces realistic 3D facial expressions in real time.

• 대한전자공학회논문지
• /
• 제26권8호
• /
• pp.1247-1256
• /
• 1989

In this paper, we propose the dynamic scene segmentation algorithm using a cross mask and edge information. This method, a combination of the conventioanl feature-based and pixel-based approaches, uses edges as features and determines moving pixels, with a cross mask centered on each edge pixel, by computing similarity measure between two consecutive image frames. With simple calcualtion the proposed method works well for image consisting of complex background or several moving objects. Also this method works satisfactorily in case of rotaitional motion.

• 한국멀티미디어학회논문지
• /
• 제8권10호
• /
• pp.1369-1382
• /
• 2005

도로 영상에서 차량 영역을 분할하는 차량 영역 분할(vehicle segmentation) 문제는 지능형 교통 시스템을 비롯한 다양한 응용 분야들에서 중요하게 사용되는 기본 연산(fundamental operation)이다. 본 연구에서는 야외의 도로 상에 설치된 CCD카메라에서 촬영된 정지 영상으로부터 차량 영역을 찾아내는 효율적인 방법을 제안한다 제안하는 방법은 입력되는 영상들을 격자 단위로 분할하여 각 격자에서의 에지 검출 결과를 대표하는 특징값(feature value)들을 통계적으로 분석한 후, 이를 바탕으로 최적해를 구한다. 전처리 과정에서는 다양한 외부 환경에서 촬영한 배경 영상들에 대해서 각 격자에서의 특징값들을 통계 처리한다. 입력된 차량 영상에서는 각 격자의 특징값이 배경 영상의 대응되는 격자에서의 특징값과 통계적으로 얼마나 오차를 보이냐에 따라, 배경 영역인지 차량 영역인지를 판단한다. 격자 별로 차량 영역에 해당하는 지를 판정한 뒤, 이 결과에 동적 프로그래밍(dynamic Programming) 기법을 이용하여 차량을 포함하는 최적의 직사각형 영역을 찾아낸다. 본 논문에서 제안하는 방법은 통계 처리와 전역 탐색 기법을 사용하므로 휴리스틱에 주로 의존하는 기존 연구들에 비해 좀더 체계적이다. 또한, 배경 영상에 대한 통계 처리는 흐리거나 맑은 등의 날씨 변화 및 바람이나 진동에 의한 카메라의 흔들림과 같은 다양한 외부 요인들이 가져올 수 있는 노이즈나 오차에 대해서도 높은 신뢰성을 보여준다. 제안하는 방법을 구현한 프로토타입 시스템은 $1280\times960$ 크기의 차량 영상들을 장당 평균 0.150초의 수행 시간에 처리하였으며, 총 270장의 다양한 노이즈를 가지는 차량 영상들에 대해 $97.03\%$의 성공률을 보였다.

• 센서학회지
• /
• 제22권4호
• /
• pp.256-261
• /
• 2013

This paper presents an extension of the dynamic range in a complementary metal-oxide-semiconductor (CMOS) active pixel sensor (APS) using a stacked photodiode and feedback structure. The proposed APS is composed of two additional MOSFETs and stacked P+/N-well/P-sub photodiodes as compared with a conventional APS. Using the proposed technique, the sensor can improve the spectral response and dynamic range. The spectral response is improved using an additional stacked P+/N-well photodiode, and the dynamic range is increased using the feedback structure. Although the size of the pixel is slightly larger than that of a conventional three-transistor APS, control of the dynamic range is much easier than that of the conventional methods using the feedback structure. The simulation and measurement results for the proposed APS demonstrate a wide dynamic range feature. The maximum dynamic range of the proposed sensor is greater than 103 dB. The designed circuit is fabricated by the $0.35-{\mu}m$ 2-poly 4-metal standard CMOS process, and its characteristics are evaluated.

• 한국정밀공학회지
• /
• 제25권2호
• /
• pp.131-139
• /
• 2008

This paper presents a new saliency map which is constructed by providing dynamic weights on individual features in an input image to search ROI(Region Of Interest) or FOA(Focus Of Attention). To construct a saliency map on there is no a priori information, three feature-maps are constructed first which emphasize orientation, color, and intensity of individual pixels, respectively. From feature-maps, conspicuity maps are generated by using the It's algorithm and their information quantities are measured in terms of entropy. Final saliency map is constructed by summing the conspicuity maps weighted with their individual entropies. The prominency of the proposed algorithm has been proved by showing that the ROIs detected by the proposed algorithm in ten different images are similar with those selected by one-hundred person's naked eyes.