• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권9호
• /
• 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.

• 대한원격탐사학회지
• /
• 제37권6_3호
• /
• pp.1931-1942
• /
• 2021

최근 고해상도 광학 위성영상의 활용성이 강조되면서 이를 이용한 지표 모니터링 연구가 활발히 수행되고 있다. 그러나 고해상도 위성영상은 낮은 시간 해상도에서 획득되기 때문에 그 활용성에 한계가 있다. 이러한 한계를 보완하기 위해 서로 다른 시간 및 공간 해상도를 갖는 다중 위성영상을 융합해 높은 시공간 해상도의 합성 영상을 생성하는 시공간 자료 융합을 적용할 수 있다. 기존 연구에서는 중저해상도의 위성영상을 대상으로 시공간 융합 모델이 개발되어 왔기 때문에 고해상도 위성영상에 대한 기개발된 융합 모델의 적용성을 평가할 필요가 있다. 이를 위해 이 연구에서는 KOMPSAT-3A 영상과 Sentinel-2 영상을 대상으로 기개발된 시공간 융합 모델의 적용성을 평가하였다. 여기에는 예측을 위해 사용하는 정보가 다른 Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM)과 Spatial Time-series Geostatistical Deconvolution/Fusion Model (STGDFM)을 적용하였다. 연구 결과, 시간적으로 연속적인 반사율 값을 결합하는 STGDFM의 예측 성능이 ESTARFM 보다 높은 것으로 나타났다. 특히 KOMPSAT 영상의 낮은 시간 해상도로 같은 시기에서 KOMPSAT 및 Sentinel-2 영상을 동시에 획득하기 어려운 경우, STGDFM의 예측 성능 향상이 더욱 크게 나타났다. 본 실험 결과를 통해 연속적인 시간 정보를 결합해 상대적으로 높은 예측 성능을 가지는 STGDFM을 이용해 낮은 재방문 주기로 인한 고해상도 위성영상의 한계를 보완할 수 있음을 확인하였다.

• 한국시스템다이내믹스연구
• /
• 제16권2호
• /
• pp.75-100
• /
• 2015

The purpose of this study is to analyze causation of Ecosystem service's trade-off(ES trade-off) and to establish baseline data for wise spatial planning and management. In order to understand why and how ES trade-off occurs, systems thinking and causal loops were employed. The causal loop of ecosystem service creation cycle includes profits quantification process, decision making process, spatial planning and management process, and ecosystem services creation process. The profits quantification process has a limitation that all ecosystem service categories were not included in profits quantification, because quantification method for cultural services is insufficient. These problems led to unequal discussion opportunity in decision making process. ES trade-off occurs through transition of ecosystem function in spatial scale and temporal scale. In spatial scale, land-use variation and resource-use variation contribute to change an ecosystem function for different ES category by spatial planning and management. In temporal scale, a change of an ecosystem function for different ES category is influenced by ecological succession, seasonal change and land cover variation, which are parameter from environmental features. This study presented that spatial planning and management should ecosystem service assessment in order to enhance balanced ecosystem services.

• 정보과학회 논문지
• /
• 제43권9호
• /
• pp.963-973
• /
• 2016

대용량 미디어 온톨로지를 이용하여 의미 있는 지능형 서비스를 제공하기 위해 기존의 Axiom 추론뿐만 아니라 다양한 추론을 활용하는 지식 확장이 요구되고 있다. 특히 시공간 정보는 인공지능 응용분야에서 중요하게 활용될 수 있고, 시공간 정보의 표현과 추론에 대한 중요도는 지속적으로 증가하고 있다. 따라서 본 논문에서는 공간 정보를 추론에 활용하기 위해서 공공 주소체계에 대한 LOD를 대용량 미디어 온톨로지에 추가하고, 이러한 대용량 데이터 처리를 위해 인메모리 기반의 분산 처리 프레임워크를 활용하는 공간 추론을 포함하는 RDFS 추론 시스템을 제안한다. 또한 추론을 통해 확장된 데이터를 포함하는 대용량 온톨로지 데이터를 대상으로 하는 분산 병렬 시공간 SPARQL 질의 처리 방법에 대해서 설명한다. 제안하는 시스템의 성능을 측정하기 온톨로지 추론과 질의 처리 벤치 마킹을 위한 LUBM과 BSBM 데이터셋을 대상으로 실험을 진행했다.

• Journal of Forest and Environmental Science
• /
• 제33권3호
• /
• pp.161-171
• /
• 2017

Global environmental changes have the capacity to make dramatic alterations to floral and faunal composition, and elucidation of the mechanism is important for predicting its outcomes. Studies on global climate change have traditionally focused on statistical summaries within relatively wide scales of spatial and temporal changes, and less attention has been paid to variability in microclimates across spatial and temporal scales. Microclimate is a suite of climatic conditions measured in local areas near the earth's surface. Environmental variables in microclimatic scale can be critical for the ecology of organisms inhabiting there. Here we examine the effect of spatial and temporal changes in microclimates on those of carabid beetle communities in Hyangnobong, Korea. We found that climatic variables and the patterns of annual changes in carabid beetle communities differed among sites even within the single mountain system. Our results indicate the importance of temporal survey of communities at local scales, which is expected to reveal an additional fraction of variation in communities and underlying processes that has been overlooked in studies of global community patterns and changes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권3호
• /
• pp.1595-1613
• /
• 2017

The IR camera and laser-based IR projector provide an effective solution for real-time collection of moving targets in RGB-D videos. Different from the traditional RGB videos, the captured depth videos are not affected by the illumination variation. In this paper, we propose a novel feature extraction framework to describe human activities based on the above optical video capturing method, namely spatial-temporal texture features for 3D human activity recognition. Spatial-temporal texture feature with depth information is insensitive to illumination and occlusions, and efficient for fine-motion description. The framework of our proposed algorithm begins with video acquisition based on laser projection, video preprocessing with visual background extraction and obtains spatial-temporal key images. Then, the texture features encoded from key images are used to generate discriminative features for human activity information. The experimental results based on the different databases and practical scenarios demonstrate the effectiveness of our proposed algorithm for the large-scale data sets.

• Journal of Electrical Engineering and Technology
• /
• 제10권4호
• /
• pp.1851-1863
• /
• 2015

– In the paper, to handle the problem of traditional bag-of-features model ignoring the spatial relationship of local features in human action recognition, we proposed a Multiscale Spatial Position Coding under Locality Constraint method. Specifically, to describe this spatial relationship, we proposed a mixed feature combining motion feature and multi-spatial-scale configuration. To utilize temporal information between features, sub spatial-temporal-volumes are built. Next, the pooled features of sub-STVs are obtained via max-pooling method. In classification stage, the Locality-Constrained Group Sparse Representation is adopted to utilize the intrinsic group information of the sub-STV features. The experimental results on the KTH, Weizmann, and UCF sports datasets show that our action recognition system outperforms the classical local ST feature-based recognition systems published recently.

• 대한원격탐사학회지
• /
• 제34권1호
• /
• pp.89-99
• /
• 2018

Spatial downscaling is often applied to coarse scale satellite products with high temporal resolution for environmental monitoring at a finer scale. An area-to-point regression kriging (ATPRK) algorithm is regarded as effective in that it combines regression modeling and residual correction with area-to-point kriging. However, an open source tool or package for ATPRK has not yet been developed. This paper describes the development and code organization of an R-based spatial downscaling tool, named R4ATPRK, for the implementation of ATPRK. R4ATPRK was developed using the R language and several R packages. A look-up table search and batch processing for computation of ATP kriging weights are employed to improve computational efficiency. An experiment on spatial downscaling of coarse scale land surface temperature products demonstrated that this tool could generate downscaling results in which overall variations in input coarse scale data were preserved and local details were also well captured. If computational efficiency can be further improved, and the tool is extended to include certain advanced procedures, R4ATPRK would be an effective tool for spatial downscaling of coarse scale satellite products.

• Wind and Structures
• /
• 제5권5호
• /
• pp.441-462
• /
• 2002

This paper presents a practical numerical method to determine both the spatial and temporal distribution of driving rain on buildings. It is based on an existing numerical simulation technique and uses the building geometry and climatic data at the building site as input. The method is applied to determine the 3D spatial and temporal distribution of wind-driven rain on the facade a low-rise building of complex geometry. Distinct wetting patterns are found. The important causes giving rise to these particular patterns are identified : (1) sweeping of raindrops towards vertical building edges, (2) sweeping of raindrops towards top edges, (3) shelter effect by various roof overhang configurations. The comparison of the numerical results with full-scale measurements in both space and time for a number of on site recorded rain events shows the numerical method to yield accurate results.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2009년도 학술발표회 초록집
• /
• pp.1260-1264
• /
• 2009

Precipitation time series is a mixture of complicate fluctuation and changes. The monthly precipitation data of 61 stations during 36 years (1973-2008) in Korea are comprehensively analyzed using the EOFs technique and CSEOFs technique respectively. The main motivation for employing this technique in the present study is to investigate the physical processes associated with the evolution of the precipitation from observation data. The twenty-five leading EOF modes account for 98.05% of the total monthly variance, and the first two modes account for 83.68% of total variation. The first mode exhibits traditional spatial pattern with annual cycle of corresponding PC time series and second mode shows strong North South gradient. In CSEOF analysis, the twenty-five leading CSEOF modes account for 98.58% of the total monthly variance, and the first two modes account for 78.69% of total variation, these first two patterns' spatial distribution show monthly spatial variation. The corresponding mode's PC time series reveals the annual cycle on a monthly time scale and long-term fluctuation and first mode's PC time series shows increasing linear trend which represents that spatial and temporal variability of first mode pattern has strengthened. Compared with the EOFs analysis, the CSEOFs analysis preferably exhibits the spatial distribution and temporal evolution characteristics and variability of Korean historical precipitation.