• IEIE Transactions on Smart Processing and Computing
• /
• 제4권2호
• /
• pp.110-114
• /
• 2015

This paper reports a novel object-detection technique in video sequences. The proposed algorithm consists of detection of objects in illuminated and occluded videos by using object features and a neural network technique. It consists of two functional modules: region-based object feature extraction and continuous detection of objects in video sequences with region features. This scheme is proposed as an enhancement of the Lowe's scale-invariant feature transform (SIFT) object detection method. This technique solved the high computation time problem of feature generation in the SIFT method. The improvement is achieved by region-based feature classification in the objects to be detected; optimal neural network-based feature reduction is presented in order to reduce the object region feature dataset with winner pixel estimation between the video frames of the video sequence. Simulation results show that the proposed scheme achieves better overall performance than other object detection techniques, and region-based feature detection is faster in comparison to other recent techniques.

• 한국멀티미디어학회논문지
• /
• 제17권4호
• /
• pp.458-465
• /
• 2014

In this paper, we propose a strategy to distribute the energy consumption over the network. The proposed strategy is based on geographic routing. We use a smart base station that maintains the residual energy and location information of sensor nodes and selects a head node and an anchor node using this information. A head node gathers and aggregates data from the sensor nodes in a target region that interests the user. An anchor node then transmits the data that was forwarded from the head node back to the smart base station. The smart base station extends network lifetime by selecting an optimal head node and an optimal anchor node. We simulate the proposed protocol and compare it with the LEACH protocol in terms of energy consumption, the number of dead nodes, and a distribution map of dead node locations.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 추계학술대회 논문집 전력기술부문
• /
• pp.76-79
• /
• 2002

In this paper, we propose a Probabilistic Tabu Search(PTS) method for hydro-thermal scheduling. Hydro scheduling has many constraints and very difficult to solve the optimal schedule because it has many local minima. To solve the problem effectively, the proposed method uses two procedures, one is Tabu search procedure that plays a role in local search, and the other is Restarting procedure that enables to diversify its search region. To adjust parameters such as a reducing rate and initial searching region, search strategy is selected according to its probability after Restarting procedure. In order to show the usefulness of the proposed method, the PTS is applied on two cases which have dependent hydro plants and compared to those of other method. The simulation results show it is very efficient and useful algorithm to solve the hydro-thermal scheduling problem.

• 대한기계학회논문집A
• /
• 제29권10호
• /
• pp.1430-1437
• /
• 2005

It is very important to achieve a high reproducibility in the ultrasonic measurement of bone mineral density. In this study, we examined number of sampling waveform, control of temperature, diameter of region of interest as factors to improve reproducibility. We decided the optimal number of waveforms to be converted to frequency domain as period of 1. We have minimized the effects of variable temperature and constrained generation of micro bubble by keeping temperature within a range of $32\pm0.5^{\circ}C$ with a precise temperature controlling algorithm. We also found the optimal diameter of region of interest to be 13mm. In this paper, we demonstrated the improved reproducibility by controlling various factors affecting the ultrasonic measurement of bone mineral density.

• Journal of Magnetics
• /
• 제22권2호
• /
• pp.234-237
• /
• 2017

This study is aimed to optimize a location of region of interest (ROI) in test bolus carotid contrast enhanced magnetic resonance angiography (CE-MRA) at 3.0T. A total of consecutive 270 patients with no cardiovascular and vessel diseases were selected. Patients underwent elliptical centric 3D CE-MRA with the test bolus technique to identify the individual arterial arrival time. Quantitative measurements were performed by drawing ROIs of $25mm^2$ and signal intensities (SI) were measured in the center of common carotid artery (CCA), internal carotid artery (ICA) and aortic arch (AA). As a result, ROIs located within AA showed a significantly clarified arterial peak and over three times increased SI, while no significant arterial peak time differences were observed compared to ROIs located within CCA. In conclusion, it was demonstrated that the aortic arch is the optimal position to locate ROI in test bolus images of the carotid CE-MRA.

• 물과 미래
• /
• 제27권3호
• /
• pp.135-143
• /
• 1994

본 논문에서는 분기형 관수로 설계방법의 모델을 제시하기 위하여 선형계획법(LP)을 도입한 설계를 연구하였다. 실제 사업지구인 전남해남군 간척사업지구의 자료를 토대로 LP 수식화에 필요한 자료를 수집하여 파이프 관경과 펌프마력설계를 최적화하였다. 연구결과 기존의 관망설계와 비교해 보았을 때 파이프 관경과 펌프마력등에서 더 경제적인 결과를 얻을 수 있었고, 수리모의모형을 사용한 기존의 설계방법보다 객관적이고 효율적인 설계가 가능했다. 이러한 결과를 바탕으로 본 논문에서 연구된 선형계획법을 이용한 분기형 관망설게의 모형이 실무에서도 효율적으로 적용될 수 있음을 알 수 있었다.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• 제24권4호
• /
• pp.331-350
• /
• 2020

Exploding gradient is a widely known problem in training recurrent neural networks. The explosion problem has often been coped with cutting off the gradient norm by some fixed value. However, this strategy, commonly referred to norm clipping, is an ad hoc approach to attenuate the explosion. In this research, we opt to view the problem from a different perspective, the discrete-time optimal control with infinite horizon for a better understanding of the problem. Through this perspective, we fathom the region at which gradient explosion occurs. Based on the analysis, we introduce a gradient-explosion-free algorithm that keeps the training process away from the region. Numerical tests show that this algorithm is at least three times faster than the clipping strategy.

• 풍력에너지저널
• /
• 제3권2호
• /
• pp.34-41
• /
• 2012

The paper is focusing on investigating the control characteristics of the baseline controller of 5 MW wind turbine provided by NREL(National Renewable Energy Laboratory). The baseline controller consist of two control logics, a maximum power tracking control below the rated wind speed and a constant power control above the rated wind speed. In the low wind speed, the mean generator power for changing the turbulent intensity and the optimal constant is studied through numerical simulations using FAST program. On the other hand, the constant power control logic and the constant control logic are compared in the high wind speed. It is confirmed that optimal constant is closely related to the turbulent intensity in low wind speed region and the constant torque control has better performance than the constant power control with respect to mechanical load in high wind speed region.

• 한국음향학회지
• /
• 제31권7호
• /
• pp.435-440
• /
• 2012

본 논문에서는 enhanced aacPlus의 Parametric Stereo(PS)에서 스테레오 파라미터의 컨텍스트 적응 코딩을 위해 최적의 컨텍스트를 제안한다. 스테레오 파라미터의 양자화 인덱스에 대해 컨텍스트 후보는 시간-스테레오 밴드도메인에서 소스 인덱스와 인접한 인덱스 값 및 이들의 조합으로 8개가 제안되었으며, 리프레시 프레임 여부 및 스테레오 밴드에 따라서 다시 4개의 상황에 대해 이들 중 사용가능한 후보들을 이용하였다. 실험에 의해 상황별로 최적의 컨텍스트가 제안되었고, 추후 이들을 이용하여 기존의 방법보다 더 좋은 성능을 가지는 PS의 코딩 방법이 가능할 것으로 기대된다.

• 한국항행학회논문지
• /
• 제11권1호
• /
• pp.105-111
• /
• 2007

본 논문에서는 퍼지추론을 이용하여 워터마크가 삽입될 최적의 부 대역을 결정하는 방법을 기술하였다. 제안한 방법은 퍼지추론 및 대조도 등과 같은 인간의 시각시스템도 고려하였다. 먼저 시각시스템과 통계적 특성을 활용하여 최적의 계수 대역을 설정하였고, 외부환경 변화에도 대응할 수 있는 퍼지 추론기를 설계 하였다. 비가시성과 워터마크의 정확도를 실험한 결과, 대조도는 스므드 영상에서 우수하였고, 러프한 영상에서는 통계적 특성이 우수함을 보였다.