• 한국정보통신학회논문지
• /
• 제11권11호
• /
• pp.2172-2178
• /
• 2007

실내 환경에서 이러한 두가지변수인 대규모에서의 경로손실과 소규모에서의 페이딩현상은 거리에 대한 RSSI(Received Signal Strength Indicator) 값의 비선형적인 변화를 유발하게 되며 이러한 현상이 실내위치 추적에서의 문제점의 하나로 지적되고 있다. 이 연구에서는 동일한 방에서의 다른 위치와 시간에서의 RSSI변화를 실험에 의한 통계에 의해 찾아서 보다 정밀한 모델을 세워서 실내 RSSI 특성화를 이루려고 하였다. 실험에서 RSSI값이 공간과 일시적인 요인 두가지에 의해 결정되는 것이 확인되었고 다른 위치에 있는 모든 센서 노드도 공간차라메터는 다르지만 임시파라메터값은 동일하다는 것을 확인하였다. 임시 파라메터들도 환경변화에 따라 천천히 신간에 따라 변화하는 대규모적인 변수의 특성을 지닌다. 이러한 관계를 활용하여 위치추적을 보다 효율적이고 정밀하게 평가할 수 있었다.

• International Journal of Contents
• /
• 제17권3호
• /
• pp.1-14
• /
• 2021

Since the outbreak of the coronavirus disease 2019 (COVID-19) pandemic, a lot of efforts have been made in the field of data science to help combat against this disease. Among them, forecasting the number of cases of infection is a crucial problem to predict the development of the pandemic. Many deep learning-based models can be applied to solve this type of time series problem. In this research, we would like to take a step forward to incorporate spatial data (geography) with time series data to forecast the cases of region-level infection simultaneously. Specifically, we model a single spatio-temporal graph, in which nodes represent the geographic regions, spatial edges represent the distance between each pair of regions, and temporal edges indicate the node features through time. We evaluate this approach in COVID-19 in a Korean dataset, and we show a decrease of approximately 10% in both RMSE and MAE, and a significant boost to the training speed compared to the baseline models. Moreover, the training efficiency allows this approach to be extended for a large-scale spatio-temporal dataset.

• Ocean and Polar Research
• /
• 제44권2호
• /
• pp.179-190
• /
• 2022

Data missing in optical satellite images caused by natural variations have been a crucial barrier in observing the status of marine surfaces. Although there have been many attempts to fill the gaps of non-observation, there is little research to analyze the ratio of missing grids to overall sea grids and their seasonal patterns. This report introduces the method of quantifying the distribution of missing points and then shows how the missing points have spatial correlation and seasonal trends. Both temporal and spatial integration methods are compared to assess the effectiveness of reducing missing data. The temporal integration shows more outstanding performance than the spatial integration. Moran's I and K-function with statistical hypothesis testing show that missing grids are clustered and there is a non-random distribution from daily integration. The result of the seasonality test for Moran's I through a periodogram shows dependency on full-year, half-year, and quarter-year periods respectively. These analysis results can be used to deduce appropriate integration periods with permissible estimation errors.

• 한국IT서비스학회지
• /
• 제15권3호
• /
• pp.211-221
• /
• 2016

The human spatial mobility information is in high demand in various businesses; however, there are only few studies on human mobility because spatio-temporal data is insufficient and difficult to collect. Now with the spread of smartphones and the advent of social networking services, the spatio-temporal data began to occur on a large scale, and the data is available to the public. In this work, we compared the movement behavior of residents and tourists by using geo-tagged tweets which contain location information. We chose Seoul to be the target area for analysis. Various creative concepts and analytical methods are used: grid map concept, cells visited concept, reverse geocoding concept, average activity index, spatial mobility index, and determination of residents and visitors based on the number of days in residence. Conducting a series of analysis, we found significant differences of the movement behavior between local residents and tourists. We also discovered differences in visiting activity according to residential countries and used applications. We expect that findings of this research can provide useful information on tourist development and urban development.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권11호
• /
• pp.5624-5638
• /
• 2016

A novel method based on Surf points is proposed to detect and lock-track single ground target in aerial videos. Videos captured by moving cameras contain complex motions, which bring difficulty in moving object detection. Our approach contains three parts: moving target template detection, search area estimation and target tracking. Global motion estimation and compensation are first made by grids-sampling Surf points selecting and matching. And then, the single ground target is detected by joint spatial-temporal information processing. The temporal process is made by calculating difference between compensated reference and current image and the spatial process is implementing morphological operations and adaptive binarization. The second part improves KALMAN filter with surf points scale information to predict target position and search area adaptively. Lastly, the local Surf points of target template are matched in this search region to realize target tracking. The long-term tracking is updated following target scaling, occlusion and large deformation. Experimental results show that the algorithm can correctly detect small moving target in dynamic scenes with complex motions. It is robust to vehicle dithering and target scale changing, rotation, especially partial occlusion or temporal complete occlusion. Comparing with traditional algorithms, our method enables real time operation, processing $520{\times}390$ frames at around 15fps.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2023년도 학술발표회
• /
• pp.27-27
• /
• 2023

To analyze the evolution of water resources in Qiandao Lake Basin under the condition of climate change, a WEP-L distributed hydrological model was established to simulate the water cycle process in the basin during 1960-2020. The Mann-Kendall non-parametric test method and Hurst index method were used to analyze the inter-annual variation and annual distribution characteristics of the total water resources in the basin. The multi-scale temporal and spatial distribution and evolution trend of water resources in Qiandao Lake Basin were evaluated. The results show that: (1) The WEP-L model has good simulation results in the Qiandao Lake basin, and the Nash coefficient rate is above 0.83 in the periodic period and above 0.85 in the verification period. (2) The water yield coefficient of the whole basin ranges from 0.436 to 0.630. The annual average total water resource is 12.25 billion m³, equivalent to 1176.4mm of water depth. The annual distribution process shows a unimodal structure, and the water depth of each sub-basin ranges from 742 mm to 1266 mm, and the spatial distribution is higher in the west and lower in the east. (3) The annual water resources series in the basin showed an insignificant upward trend, and the Hurst index was 0.86, indicating a continuous upward trend. From the perspective of monthly water resources, January and February increased significantly, the other months were not significant changes.

• Wind and Structures
• /
• 제29권6호
• /
• pp.471-488
• /
• 2019

Wind pressure is a critical argument for the wind-resistant design of structures. The attempt, however, to explore the wind pressure field on buildings still encounters challenges though a large body of researches utilizing wind tunnel tests and wind field simulations were carried out, due to the difficulty in logical treatments on the scale effect and the modeling error. The full-scale measurement has not yet received sufficient attention. By performing a field measurement, the present paper systematically addresses wind pressures on the rectangular attic of a double-tower building. The spatial and temporal correlations among wind speed and wind pressures at measured points are discussed. In order to better understand the wind pressure distribution on the attic facades and its relationship against the approaching flow, a full-scale CFD simulation on the similar rectangular attic is conducted as well. Comparative studies between wind pressure coefficients and those provided in wind-load codes are carried out. It is revealed that in the case of wind attack angle being zero, the wind pressure coefficient of the cross-wind facades exposes remarkable variations along both horizontal and vertical directions; while the wind pressure coefficient of the windward facade remains stable along horizontal direction but exposes remarkable variations along vertical direction. The pattern of wind pressure coefficients, however, is not properly described in the existing wind-load codes.

• 한국인터넷방송통신학회논문지
• /
• 제23권6호
• /
• pp.89-95
• /
• 2023

인간 행동 인식 (Human action recognition, HAR) 기술은 스포츠 분석, 인간과 로봇 간의 상호작용, 대형 사이니지 콘텐츠 등의 애플리케이션에 활용되는 핵심 기술 중 하나이다. 본 논문에서는 몰입형 대형 사이니지 콘텐츠를 위한 STAGCN (Spatial temporal attention graph convolutional network) 기반 인간 행동 인식 시스템을 제안한다. STAGCN은 attention mechanism을 통해 스켈레톤 시퀀스의 시공간적 특징에 서로 다른 가중치를 부과하여, 동작 인식에 중요한 관절 및 시점을 고려할 수 있다. NTU RGB+D 데이터셋을 사용한 실험 결과, 제안된 시스템은 기존 딥러닝 모델들에 비해 높은 분류 정확도를 달성한 것을 확인했다.

• 한국수자원학회논문집
• /
• 제49권9호
• /
• pp.761-771
• /
• 2016

기상청에서 운영하고 제공하는 전지구 계절예측시스템 GloSea5 (Global Seasonal forecasting system version 5)자료를 활용하여 용담댐유역에 적용하고자 하였다. GloSea5는 예측자료(Forecast; 이하 FCST)와 과거재현자료(Hindcast; 이하 HCST)로 제공되며 공간 수평해상도는 N216 ($0.83^{\circ}{\times}0.56^{\circ}$)으로 중위도에서 약 60km이다. 이를 유역단위 물관리에 활용하기 위해서는 시 공간적인 상세화가 필요하므로 통계적 상세화 기법을 수행하여 변수가 갖는 계통적인 지역 오차를 보정함으로써 자료의 신뢰도를 향상시키고자 하였다. HCST자료는 앙상블 형태로 주어지며 용담댐 유역의 앙상블 평균에 대한 6번 격자의 통계적인 상관성($R^2=0.60$, RMSE=88.92, NSE=0.57)이 가장 높게 나타났다. 또한 계절분석시 여름철의 경우 원시 GloSea5 강우량이 600.1mm로 관측값인 816.1mm 대비 -26.5%로 가장 많은 차이를 보였으며 상세화 후 GloSea5 강우량은 -3.1%의 오차율을 보였다. 대부분의 과소 모의된 결과가 여름철 홍수기에 해당되는 강우로 상세화 이후 강우가 회복되는 매우 중요한 결과를 보였다. 계절별 Moran's I 지수를 이용한 공간적 자기상관분석 결과 역시 통계적으로 유의성 있는 공간적인 분포를 나타냄으로써 자료의 불확실성을 개선하고 시 공간적인 정확도와 타당성을 입증하였다. HCST기간에 대한 GloSea5의 앙상블 강우에 대한 신뢰도를 향상시킴으로써 수문학적인 영향을 평가하기 위한 자료로서의 충분한 가능성을 확보하였으며 이러한 시 공간적인 재현성에 대한 평가결과는 향후 유역단위 물관리를 위한 기초자료로서 매우 중요한 역할을 할 것이다.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권12호
• /
• pp.1075-1082
• /
• 2007

본 논문은 동영상에 등장하는 다수 사람의 동작을 검출하여 검출된 동작을 개별적으로 인식하는 방법을 제안한다. 동작이 수행되는 속도 또는 크기 변화에 강인한 인식 성능을 갖기 위해 시공간축 피라미드(Spatial-Temporal Pyramid)방식을 적용한다. 동작 표현 방식을 통계적 특성 기반의 모션 그래디언트 히스토그램(MGH:Motion Gradient Histogram)으로 선택하여 인식 과정에서 발생하는 복잡도를 최소화 하였다. 다수의 동작을 검출하기 위하여 이진 차영상을 축적한 모션 에너지 이미지(MEI: Motion Energy Image) 방법을 적용하여 효율적으로 개별적 동작 영역을 획득한다. 각 영역은 동작 표현 방법인 MGH로 나타내어지고, 크기 변화에 강인하도록 피라미드 방식을 적응하여 학습된 템플릿 MGH와 유사도를 상호 비교하여 최종 인식 결과를 얻는다. 인식 성능의 평가를 위해 10개의 동영상을 활용하여 단일 객체, 다수 객체, 속도 및 크기 변화, 기존 방식과의 비교, 기타 추가 실험 등을 실시하여 다양한 조건의 영상에서 양호한 인식 결과를 확인 할 수 있었다.