• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.180-180
• /
• 2015

Weather radar has been widely used in measuring precipitation and discharge and predicting flood risks. The radar rainfall estimate has one of the essential problems in terms of uncertainty and accuracy. Previous study analyzed radar errors to reduce its uncertainty or to improve its accuracy. Furthermore, a recent analyzed the effect of radar error on rainfall-runoff using spatial error model (SEM). SEM appropriately reproduced radar error including spatial correlation. Since the SEM does not take the time dependence into account, its time variability was not properly investigated. Therefore, in the current study, we extend the SEM including time dependence as well as spatial dependence, named after Spatial-Temporal Error Model (STEM). Radar rainfall events generated with STEM were tested so that the peak runoff from the response of a basin could be investigated according to dependent error. The Nam River basin, South Korea, was employed to illustrate the effects of STEM on runoff peak flow.

• Journal of KIISE:Databases
• /
• v.27 no.1
• /
• pp.101-112
• /
• 2000

In Geographic Information System, spatial object can be changed in the attribute information, spatial location and the topological relation between them with the change of time. However traditional GIS deletes the old value of aspatial information and replaces them with new value. Therefore. it is difficult to manage thc history of changed spatial object and can not support the spatio-temporal queries including temporal queries. In this paper, we propose a spatio-temporal objected model to solve this problem. We implement the proposed model with spatio-temporal class using Gothic GIS tool. The historical information of an object is stored into the object itself for the effective history management. And, in order to provide the queries for the history of an object and spatio-temporal relationship, we add temporal operators, spatio-temporal operators, and spatio-temporal query operations into Gothic, and improve the facility of the Gothic.

• Journal of Information Processing Systems
• /
• v.15 no.2
• /
• pp.399-409
• /
• 2019

Three-dimensional (3D) human pose reconstruction from single-view image is a difficult and challenging topic. Existing approaches mostly process frame-by-frame independently while inter-frames are highly correlated in a sequence. In contrast, we introduce a novel spatial-temporal 3D human pose reconstruction framework that leverages both intra and inter-frame relationships in consecutive 2D pose sequences. Orthogonal matching pursuit (OMP) algorithm, pre-trained pose-angle limits and temporal models have been implemented. Several quantitative comparisons between our proposed framework and recent works have been studied on CMU motion capture dataset and Vietnamese traditional dance sequences. Our framework outperforms others by 10% lower of Euclidean reconstruction error and more robust against Gaussian noise. Additionally, it is also important to mention that our reconstructed 3D pose sequences are more natural and smoother than others.

• Spatial Information Research
• /
• v.7 no.1
• /
• pp.39-48
• /
• 1999

A GRId-based Soil MOsture Routing Model(GRISMORM) which predicts temporal variation and spatial distribution of water balance on a daily time step for each grid element of the watershed was developed. The model was programmed by C-language which aims for high flexibility to any kind of GIS softwares. The model uses ASCII-formatted map data supported by the irregular gridded map of the GRASS(Geographic Resources Analysis Support System)-GIS and generates daily or monthly spatial distribution map of water balance components within the watershed. The model was applied to Ipyunggyo watershed(75.6$km^2$) ; the part of Bocheongchun watershed. Predicted streamflows resulting from two years(95 and 96) daily data were compared with those observed at the watershed outlet. The results of temporal variation and spatial distribution of soil moisture are also presented by using GRASS.

• Journal of the Korean association of regional geographers
• /
• v.18 no.4
• /
• pp.414-432
• /
• 2012

Land price is a kind of text to read urban spatial structure. The purpose of this paper is to inquire into the characteristics of Daegu's urban structure and its change in time through exploring spatio-temporal variations of land price with a detailed spatial and temporal resolution. To achieve this, land value surfaces were represented using the officially assessed land price every other year from 1995 to 2011. Through mapping and exploring spatio-temporal patterns and fluctuation rates of land price for this period, changes in urban structure, the effects of local decision makings such as Greenbelt adjustment, housing site development, and gentrification, and the effects of business fluctuations or policies at global or national scales could be caught. In addition, the trends for suburbanization and multi-centric urban form could be examined from the results of a negative exponential model explaining the effect of distance from an urban center on spatial variation of land price. These results demonstrate that urban analysis using land price mirroring spatial decision making at various scales could deepen understanding for internal structure and change of a city and provide useful information for establishing regional and urban development policies and evaluating their effects.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.7 no.4
• /
• pp.155-165
• /
• 2004

Hydrological model which is proper to watershed characteristics and analysis purpose must be used when we analyze water resources. But, although proper model is used, if objectivity and reasonability of data is low it is difficult to get good results from the model. So it is very important to decide the data which is used in selected model and estimate parameters by using the applied data. In this study, temporal and spatial data was constructed as standard data of test site and stored in HyGIS (Hydrological Geographic Information System) DB. A system which extracts temporal and spatial data required to run hydrological model from HyGIS DB by connecting TOPMODEL with HyGIS was developed. In this system, we can extract temporal and spatial data which is needed to run TOPMODEL from HyGIS DB and estimate model parameters by using genetic algorithm. We found that HyGIS and the system connected with TOPMODEL was effective to make temporal and spatial data used in TOPMODEL and estimate model parameters. From this study, we suggested the possibility that HyGIS could be applied properly to another hydrological model, too.

• The Journal of Korean Association of Computer Education
• /
• v.10 no.2
• /
• pp.101-113
• /
• 2007

In this paper, we present an object-based modeling and language for an object-oriented spatio-temporal database system. For handling the structure of spatio-temporal objects and the spatio-temporal operators, we propose the two layers of data modeling: a spatio-temporal object model (STOM) and an spatio_temporal internal description model (STIM). We then propose STOQL, a spatio-temporal object-oriented query language. STOQL provides an integrated mechanism for the graphical display of spatial objects and the retrieval of spatio-temporal and aspatial objects.

• Atmosphere
• /
• v.20 no.4
• /
• pp.495-503
• /
• 2010

Satellite data have an intrinsic problem due to a number of various physical parameters, which can have a similar effect on measured radiance. Most evaluations of satellite performance have relied on comparisons with limited spatial and temporal resolution of ground-based measurements such as soundings and in-situ measurements. In order to overcome this problem, a new way of satellite data evaluation is suggested with statistical tools such as empirical orthogonal function(EOF), and singular value decomposition(SVD). The EOF analyses with OMI and OMI HCHO over northeast Asia show that the spatial pattern show high correlation with population density. This suggests that human activity is a major source of as well as HCHO over this region. However, this analysis is contradictory to the previous finding with GOME HCHO that biogenic activity is the main driving mechanism(Fu et al., 2007). To verify the source of HCHO over this region, we performed the EOF analyses with vegetation and HCHO distribution. The results showed no coherence in the spatial and temporal pattern between two factors. Rather, the additional SVD analysis between $NO_2$ and HCHO shows consistency in spatial and temporal coherence. This outcome suggests that the anthropogenic emission is the main source of HCHO over the region. We speculate that the previous study appears to be due to low temporal and spatial resolution of GOME measurements or uncertainty in model input data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.9
• /
• pp.2361-2376
• /
• 2023

The electrocardiogram (ECG) signal is commonly used to screen and diagnose cardiovascular diseases. In recent years, deep neural networks have been regarded as an effective way for automatic ECG disease diagnosis. The convolutional neural network is widely used for ECG signal extraction because it can obtain different levels of information. However, most previous studies adopt single scale convolution filters to extract ECG signal features, ignoring the complementarity between ECG signal features of different scales. In the paper, we propose a dual-scale network with convolution filters of different sizes for 12-lead ECG classification. Our model can extract and fuse ECG signal features of different scales. In addition, different spatial and time periods of the feature map obtained from the 12-lead ECG may have different contributions to ECG classification. Therefore, we add a spatial-temporal attention to each scale sub-network to emphasize the representative local spatial and temporal features. Our approach is evaluated on PTB-XL dataset and achieves 0.9307, 0.8152, and 89.11 on macro-averaged ROC-AUC score, a maximum F1 score, and mean accuracy, respectively. The experiment results have proven that our approach outperforms the baselines.

• Journal of Information Processing Systems
• /
• v.16 no.3
• /
• pp.612-628
• /
• 2020

This paper presents a new approach for detecting abnormal behaviors in complex surveillance scenes where anomalies are subtle and difficult to distinguish due to the intricate correlations among multiple objects' behaviors. Specifically, a cascaded probabilistic topic model was put forward for learning the spatial context of local behavior and the temporal context of global behavior in two different stages. In the first stage of topic modeling, unlike the existing approaches using either optical flows or complete trajectories, spatio-temporal correlations between the trajectory fragments in video clips were modeled by the latent Dirichlet allocation (LDA) topic model based on Markov random fields to obtain the spatial context of local behavior in each video clip. The local behavior topic categories were then obtained by exploiting the spectral clustering algorithm. Based on the construction of a dictionary through the process of local behavior topic clustering, the second phase of the LDA topic model learns the correlations of global behaviors and temporal context. In particular, an abnormal behavior recognition method was developed based on the learned spatio-temporal context of behaviors. The specific identification method adopts a top-down strategy and consists of two stages: anomaly recognition of video clip and anomalous behavior recognition within each video clip. Evaluation was performed using the validity of spatio-temporal context learning for local behavior topics and abnormal behavior recognition. Furthermore, the performance of the proposed approach in abnormal behavior recognition improved effectively and significantly in complex surveillance scenes.