• 한국철도학회논문집
• /
• 제8권6호
• /
• pp.559-565
• /
• 2005

In this paper, we propose a stereoscopic video coding scheme for subway accident monitoring system. The proposed designed for providing flexible video among various displays, such ass control center, station employees and train driver. We uses MPEG-2 standard for coding the left-view sequence and IBMDC coding scheme predicts matching block by interpolating both motion and disparity predicted macroblocks. To provide efficient stereoscopic video service, we define both temporally and spatially scalable layers for each eye's-view by using the concept of Spatio-Temporal scalability. The experimental results show the efficiency of proposed coding scheme by comparison with already known methods and the advantages of disparity estimation in terms of scalability overhead. According to the experimental results, we expect the proposed functionalities will play a key role in establishing highly flexible stereoscopic video codec for ubiquitous display environment where devices and network connections are heterogeneous.

• International Journal of Internet, Broadcasting and Communication
• /
• 제15권3호
• /
• pp.43-51
• /
• 2023

Slotted ALOHA (S-ALOHA) is a classical medium access control protocol widely used in multiple access communication networks, supporting distributed random access without the need for a central controller. Although stability and delay have been extensively studied in existing works, most of these studies have assumed ideal channel conditions or independent fading, and the impact of time-correlated wireless channels has been less addressed. In this paper, we investigate the queueing delay performance in S-ALOHA networks under time-correlated channel conditions by utilizing a Gilbert-Elliott model. Through simulation studies, we demonstrate how temporal correlation in the wireless channel affects the queueing delay performance. We find that stronger temporal correlation leads to increased variability in queue length, a larger probability of having queue overflows, and higher congestion levels in the S-ALOHA network. Consequently, there is an increase in the average queueing delay, even under a light traffic load. With these findings, we provide valuable insights into the queueing delay performance of S-ALOHA networks, supplementing the existing understanding of delay in S-ALOHA networks.

• 인터넷정보학회논문지
• /
• 제21권5호
• /
• pp.67-73
• /
• 2020

In this paper, we present a CNN-based gesture recognition approach which reduces the memory burden of input data. Most of the neural network-based gesture recognition methods have used a sequence of frame images as input data, which cause a memory burden problem. We use a motion history image in order to define a meaningful gesture. The motion history image is a grayscale image into which the temporal motion information is collapsed by synthesizing silhouette images of a user during the period of one meaningful gesture. In this paper, we first summarize the previous traditional approaches and neural network-based approaches for gesture recognition. Then we explain the data preprocessing procedure for making the motion history image and the neural network architecture with three convolution layers for recognizing the meaningful gestures. In the experiments, we trained five types of gestures, namely those for charging power, shooting left, shooting right, kicking left, and kicking right. The accuracy of gesture recognition was measured by adjusting the number of filters in each layer in the proposed network. We use a grayscale image with 240 × 320 resolution which defines one meaningful gesture and achieved a gesture recognition accuracy of 98.24%.

• 한국위성정보통신학회논문지
• /
• 제9권4호
• /
• pp.26-31
• /
• 2014

본 논문은 컬러 영상에서 색상과 움직임 정보를 이용하여 후보영역을 특정하고 연기의 특성들을 추출하여 신경망을 사용한 검출의 성능을 분석하여 제시한다. 기존 연기 검출 알고리즘에서는 연기의 움직임, 색상을 분석하여 후보영역으로 특정하고 그 영역 안에서 연기의 여러 특성을 분석 하는 방법을 이용한다. 하지만 대부분 처음 발생하는 연기의 색상조건을 고려하지 않았기 때문에 조기 검출에는 적절하지 못하다. 본 논문은 연기의 색상과 움직임의 특성을 분석하여 그에 알맞은 방법을 적용하여 후보영역을 폭넓게 결정하고 그 영역 내에서 연기의 확산과 투명성을 인공신경망에 적용시킴으로써 나오는 성능을 분석하였다. 모의실험 결과는 91.31%의 검출율과, 2.62%의 오검출율 성능을 확인할 수 있었다.

• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.556-570
• /
• 2021

The existing video expression recognition methods mainly focus on the spatial feature extraction of video expression images, but tend to ignore the dynamic features of video sequences. To solve this problem, a multi-mode convolution neural network method is proposed to effectively improve the performance of facial expression recognition in video. Firstly, OpenFace 2.0 is used to detect face images in video, and two deep convolution neural networks are used to extract spatiotemporal expression features. Furthermore, spatial convolution neural network is used to extract the spatial information features of each static expression image, and the dynamic information feature is extracted from the optical flow information of multiple expression images based on temporal convolution neural network. Then, the spatiotemporal features learned by the two deep convolution neural networks are fused by multiplication. Finally, the fused features are input into support vector machine to realize the facial expression classification. Experimental results show that the recognition accuracy of the proposed method can reach 64.57% and 60.89%, respectively on RML and Baum-ls datasets. It is better than that of other contrast methods.

• 한국정보과학회논문지:시스템및이론
• /
• 제26권11호
• /
• pp.1333-1343
• /
• 1999

경성 실시간 시스템의 태스크들은 논리적으로 올바른 결과를 산출해야 하지만 또한 각자의 시간 제한 조건을 만족하여야 한다. 간격제한 스케줄링은 시간 제한 조건이 시간 간격 제한으로 주어지는 실시간 태스크들을 스케줄하기 위하여 도입되었다. 간격제한 스케줄링에서의 각 태스크들은 시간 간격 제한 조건을 갖는데, 이것은 태스크의 두 연속적인 수행의 종료시간에 대해 제한을 가한다. 다시 말해, 간격제한 스케줄링에서의 각 태스크 수행은 그 태스크의 직전 수행 완료 시간으로부터 발생하는 데드라인을 갖는다. 간격제한 태스크 스케줄링에 관한 많은 연구는 단순화 방법에 기초하고 있다. 그러나, 우리는 이 논문에서 단순화 방법을 사용하지 않고, 정적 우선순위 및 정적 분리 제한 정책을 채용한 새로운 간격제한 태스크 스케줄링 방법을 제안한다. 제안된 정적 할당 방법은 스케줄링 분석 및 구현을 매우 간단히 할 수 있으며, 또한 스케줄러의 실행시간 오버헤드를 줄일 수 있다.Abstract Tasks in hard real-time systems must not only be logically correct but also meet their timing constraints. The distance-constrained scheduling has been introduced to schedule real-time tasks whose timing constraints are characterized by temporal distance constraints. Each task in the distance-constrained scheduling has a temporal distance constraint which imposes restriction on the finishing times of two consecutive executions of the task. Thus, each execution of a task in the distance-constrained scheduling has a deadline relative to the finishing time of the previous execution of the task.Much work on the distance-constrained task scheduling has been based on the reduction technique. In this paper, we propose a new scheme for the distance-constrained task scheduling which does not use the reduction technique but adopts static priority and static separation constraint assignment policy. We show that our static assignment approach can simplify the scheduling analysis and its implementation, and can also reduce the run-time overhead of the scheduler.

• 한국정보통신학회논문지
• /
• 제9권1호
• /
• pp.108-116
• /
• 2005

분산 멀티미디어 환경에서 교재 제시를 위한 미디어내${\cdot}$미디어간 동기화는 전송 데이터의 지연과 손실 및 미디어별 전송 특성 등과 같이 비동기 오차로 인해 미디어들간의 정확한 시간 관계성을 보장하기가 매우 어렵다. 미디어내 동기화는 표현률에 영향을 받기 때문에 네트워크 지연으로 인한 미디어 도착률이 비정상적일 때 지터 현상이 발생한다. 그리고 각각의 미디어 스트림에서 지터의 누적된 결과로 미디어간 스큐가 발생한다. 이런 현상은 원거리 교육에서 네트워크 지연으로 인해 학습자의 내용 인식에 혼란을 야기시키고, 송${\cdot}$수신측의 효율적인 상호 작용을 제공하지 못하고 있다. 따라서 본 논문에서는 미디어 또는 하나 이상의 미디어 사이에서 시간 관계성의 요구 사항을 유지함으로써 네트워크 지연으로 인한 문제점을 해결할 수 있다. 또한 표현률에 영향을 받는 미디어내 동기화 방법을 제안하고, 원격 교육을 위한 교재 제시 시스템을 구현하였다.

• Ocean and Polar Research
• /
• 제27권2호
• /
• pp.189-195
• /
• 2005

Several models predict large and potentially abrupt ocean circulation changes due to anthropogenic greenhouse-gas emissions. These circulation changes drive-in the models-considerable oceanic oxygen trend. A sound estimate of the observed oxygen trends can hence be a powerful tool to constrain predictions of future changes in oceanic deepwater formation, heat and carbon dioxide uptake. Estimating decadal scale oxygen trends is, however, a nontrivial task and previous studies have come to contradicting conclusions. One key potential problem is that changes in the historical observation network might introduce considerable errors. Here we estimate the likely magnitude of these errors for a subset of the available observations in the Southern Ocean. We test three common data analysis methods south of Australia and focus on the decadal-scale trends between the 1970's and the 1990's. Specifically, we estimate errors due to sparsely sampled observations using a known signal (the time invariant, temporally averaged, World Ocean Atlas 2001) as a negative control. The crossover analysis and the objective analysis methods are for less prone to spatial sampling location biases than the area averaging method. Subject to numerous caveats, we find that errors due to sparse sampling for the area averaging method are on the order of several micro-moles $kg^{-1}$. for the crossover and the objective analysis method, these errors are much smaller. For the analyzed example, the biases due to changes in the spatial design of the historical observation network are relatively small compared to the tends predicted by many model simulations. This raises the possibility to use historic oxygen trends to constrain model simulations, even in sparsely sampled ocean basins.

• 정보처리학회논문지D
• /
• 제13D권3호
• /
• pp.337-346
• /
• 2006

이동 객체들의 행동을 분석하기 위해서는 그 이동 객체 궤적들 간의 유사성 측정을 정의하는 것이 필요하다. 비록 유클리디언 공간 상에서 이동 객체들의 유사한 궤적들을 검색하는 몇몇 연구가 있었지만, 도로 네트워크 공간 상에서 이동 객체들의 궤적에 대한 유사성 연구는 거의 관심을 끌지 못 했다. 실제 응용에서, 대부분의 이동 객체들은 유클리디언 공간보다 도로 네트워크 공간 상에 존재한다. 그러나 궤적들 간의 유사성 측정에서, 기존의 방법들은 유클리디언 거리를 기반으로 하고 있으며, 오직 공간적 유사성만을 고려하고 있다. 본 논문에서, 우리는 도로네트워크 상에서 POI와 TOI를 기반으로 유사성 측정을 정의한다. 이러한 정의를 바탕으로, 우리는 궤적들 간의 시공간 유사성을 사용하여 유사한 궤적을 검색하는 방법들을 제시한다. 그리고 유사한 궤적들에 대하여 클러스터링을 수행한 결과를 보인다. 실험 결과는 각 방법에 의해 검색된 유사한 궤적들과 그 검색된 궤적 결과들이 각 방법 간에 얼마나 일치하는지를 보인다.

• 한국멀티미디어학회논문지
• /
• 제24권5호
• /
• pp.684-694
• /
• 2021

Resting-state Functional Magnetic Resonance Imaging(fMRI) data detects the temporal correlations in Blood Oxygen Level Dependent(BOLD) signal and these temporal correlations are regarded to reflect intrinsic cortical connectivity, which is deactivated during attention demanding, non-self referential tasks, called Default Mode Network(DMN). The relationship between fMRI and anatomical connectivity has not been studied in detail, however, the preceded studies have tried to clarify this relationship using Diffusion Tensor Imaging(DTI) and fMRI. These studies use method that fMRI data assists DTI data or vice versa and it is used as guider to perform DTI tractography on the brain image. In this study, we hypothesized that functional connectivity in resting state would reflect anatomical connectivity of DMN and the combined images include information of fMRI and DTI showed visible connection between brain regions related in DMN. In the previous study, functional connectivity was determined by subjective region of interest method. However, in this study, functional connectivity was determined by objective and advanced method through Independent Component Analysis. There was a stronger connection between Posterior Congulate Cortex(PCC) and PHG(Parahippocampa Gyrus) than Anterior Cingulate Cortex(ACC) and PCC. This technique might be used in several clinical field and will be the basis for future studies related to aging and the brain diseases, which are needed to be translated not only functional connectivity, but structural connectivity.