• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권9호
• /
• pp.3823-3840
• /
• 2020

Recently, auto-encoder has emerged as the most popular method in convolutional neural network (CNN) based image compression and has achieved impressive performance. In the traditional auto-encoder based image compression model, the encoder simply sends the features of last layer to the decoder, which cannot allocate bits over different spatial regions in an efficient way. Besides, these methods do not fully exploit the contextual information under different receptive fields for better reconstruction performance. In this paper, to solve these issues, a novel auto-encoder model is designed for image compression, which can effectively transmit the hierarchical features of the encoder to the decoder. Specifically, we first propose an adaptive bit-allocation strategy, which can adaptively select an importance channel. Then, we conduct the multiply operation on the generated importance mask and the features of the last layer in our proposed encoder to achieve efficient bit allocation. Moreover, we present an additional novel perceptual loss function for more accurate image details. Extensive experiments demonstrated that the proposed model can achieve significant superiority compared with JPEG and JPEG2000 both in both subjective and objective quality. Besides, our model shows better performance than the state-of-the-art convolutional neural network (CNN)-based image compression methods in terms of PSNR.

• 한국재난정보학회 논문집
• /
• 제18권1호
• /
• pp.51-62
• /
• 2022

연구목적: 이 연구는 지진 옥외 대피장소를 대상으로 생활인구(유동인구)를 측면에서 공간적 접근성을 분석한 후, 접근성 취약지역에 추가 신규 대피장소를 확충할 경우의 모의분석을 하여 개선효과를 비교함으로써 시사점을 도출하는 것을 목적으로 한다. 연구방법: GIS 네트워크분석 기반의 최적화 모형인 입지배분모형을 적용하여 접근성을 분석하고 취약지역을 식별하였다. 입지배분방법은 일정한 시간 내에 신속한 이동이 중요한 대피시설의 성격에 비추어 'Maximize Coverage(수요영역 최대화)' 방법을 적용하였고, 대피를 위한 한계 거리 및 시간기준은 보행속도를 고려하여 500m(7.5분), 1,000m(15분), 1,500m(22.5분)의 세 가지로 구분하여 분석하였다. 사례분석 대상지역은 지진 발생 시 신속한 대피와 일시적인 체류를 위한 옥외 대피장소의 기능을 고려하여 거주인구에 비해 생활인구가 월등히 많고 대피장소로 활용할 수 있는 가용공간이 크게 부족한 고밀 도심지역으로서 서울 중구지역을 선정하였다. 연구결과: 분석 결과, 전반적으로 거주 인구에 비해 생활인구 기준으로 볼 때 접근성이 취약한 집계구 수와 인구 수가 훨씬 많고 비율도 높았으며, 접근성 취약지역에 가용한 신규 대피장소를 확충할 경우의 모의분석에서 접근성이 크게 개선됨을 확인할 수 있었다. 다만, 고밀 도심지역으로서 가용지가 절대적으로 부족한 대상지역의 특성상 잠재적인 대피인구 전체의 접근성을 완전히 해소하지는 못하였다. 결론: 유동인구로 인해 주·야간 인구 차이가 심한 서울 도심의 지역 특성을 반영하기 위하여 생활인구 첨두시간대를 적용하여 실제 대피수요를 고려할 필요가 있으며, 입지배분모형을 이용하여 접근성이 불리한 취약지역을 식별하고 신규 대피장소 설치의 우선순위를 부여함으로써 과학적 근거 기반의 의사결정이 필요하다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권12호
• /
• pp.3274-3292
• /
• 2013

Unsupervised methods for image segmentation are recently drawing attention because most images do not have labels or tags. A topic model is such an unsupervised probabilistic method that captures latent aspects of data, where each latent aspect, or a topic, is associated with one homogeneous region. The results of topic models, however, usually have noises, which decreases the overall segmentation performance. In this paper, to improve the performance of image segmentation using topic models, we propose two topic masks applicable to topic assignments of homogeneous regions obtained from topic models. The topic masks capture the noises among the assigned topic assignments or topic labels, and remove the noises by replacements, just like image masks for pixels. However, as the nature of topic assignments is different from image pixels, the topic masks have properties that are different from the existing image masks for pixels. There are two contributions of this paper. First, the topic masks can be used to reduce the noises of topic assignments obtained from topic models for image segmentation tasks. Second, we test the effectiveness of the topic masks by applying them to segmented images obtained from the Latent Dirichlet Allocation model and the Spatial Latent Dirichlet Allocation model upon the MSRC image dataset. The empirical results show that one of the masks successfully reduces the topic noises.

• Asian Journal of Atmospheric Environment
• /
• 제2권1호
• /
• pp.34-46
• /
• 2008

Emissions inputs for use in air quality modeling of Korea were generated with the emissions inventory data from the National Institute of Environmental Research (NIER), maintained under the Clean Air Policy Support System (CAPSS) database. Source Classification Codes (SCC) in the Korea emissions inventory were adapted to use with the U.S. EPA's Sparse Matrix Operator Kernel Emissions (SMOKE) by finding the best-matching SMOKE default SCCs for the chemical speciation and temporal allocation. A set of 19 surrogate spatial allocation factors for South Korea were developed utilizing the Multi-scale Integrated Modeling System (MIMS) Spatial Allocator and Korean GIS databases. The mobile and area source emissions data, after temporal allocation, show typical sinusoidal diurnal variations with high peaks during daytime, while point source emissions show weak diurnal variations. The model-ready emissions are speciated for the carbon bond version 4 (CB-4) chemical mechanism. Volatile organic carbon (VOC) emissions from painting related industries in area source category significantly contribute to TOL (Toluene) and XYL (Xylene) emissions. ETH (Ethylene) emissions are largely contributed from point industrial incineration facilities and various mobile sources. On the other hand, a large portion of OLE (Olefin) emissions are speciated from mobile sources in addition to those contributed by the polypropylene industry in point source. It was found that FORM (Formaldehyde) is mostly emitted from petroleum industry and heavy duty diesel vehicles. Chemical speciation of PM2.5 emissions shows that PEC (primary fine elemental carbon) and POA (primary fine organic aerosol) are the most abundant species from diesel and gasoline vehicles. To reduce uncertainties in processing the Korea emission inventory due to the mapping of Korean SCCs to those of U.S., it would be practical to develop and use domestic source profiles for the top 10 SCCs for area and point sources and top 5 SCCs for on-road mobile sources when VOC emissions from the sources are more than 90% of the total.

• 지적과 국토정보
• /
• 제50권2호
• /
• pp.101-115
• /
• 2020

자율주행 분야 글로벌 동향 파악 및 공간정보 서비스 활성화 방안 도출을 위해 빅데이터 분석방법을 활용하였다. 사용된 빅데이터는 뉴스기사와 특허문헌을 상호 연계하여 활용하고, 뉴스 기사를 통한 동향 분석, 특허문헌 정보를 활용한 기술 분석이 진행 되었다. 본 논문에서는 자율주행에 대한 주요 뉴스에서 토픽모델을 기반으로 한 LDA(Latent Dirichlet Allocation)를 활용하여 빅데이터화 하고 주요 단어를 추출하였다. 특허정보의 주요 단어를 기반으로 적용된 워드넷(WordNet)을 활용하여 공간정보와 연계성 분석, 글로벌 기술 동향 분석을 실시하고 공간정보 분야의 동향 분석 및 예측을 실시하였다. 본 논문에서는 주요뉴스와 특허문헌 정보를 기반으로 한 빅데이터 분석방법으로 자율주행 분야와 공간정보와의 연계성 분석을 통하여 최신 동향과 미래를 예측하는 방법을 제시한다. 빅데이터 분석으로 도출된 자율주행 분야 공간정보의 글로벌 동향은 플랫폼 얼라이언스, 비지니스 파트너쉽, 기업 인수합병, 합작회사 설립, 표준화 및 기술개발로 도출되었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권9호
• /
• pp.3598-3614
• /
• 2020

With the increase of motor vehicles and tourism demand, some traffic problems gradually appear, such as traffic congestion, safety accidents and insufficient allocation of traffic resources. Facing these challenges, a model of Spatio-Temporal Dilated Convolutional Network (STDGCN) is proposed for assistance of extracting highly nonlinear and complex characteristics to accurately predict the future traffic flow. In particular, we model the traffic as undirected graphs, on which graph convolutions are built to extract spatial feature informations. Furthermore, a dilated convolution is deployed into graph convolution for capturing multi-scale contextual messages. The proposed STDGCN integrates the dilated convolution into the graph convolution, which realizes the extraction of the spatial and temporal characteristics of traffic flow data, as well as features of road occupancy. To observe the performance of the proposed model, we compare with it with four rivals. We also employ four indicators for evaluation. The experimental results show STDGCN's effectiveness. The prediction accuracy is improved by 17% in comparison with the traditional prediction methods on various real-world traffic datasets.

• 교육시설
• /
• 제16권5호
• /
• pp.11-18
• /
• 2009

This study examines the issue of space management of university facilities by an assessment model as part of efforts to deal with the crisis of universities. To this end, the study addressed efficiency issue and structural problems using assessment model factors, with the aim of figuring out legitimacy and allocating methods for this purpose. Selected model factors included utilization ratio, residual ratio, vacancy ratio, and occupancy ratio, while for the latter, we investigated into the present situations of space use, focusing on construction, design, and living dimensions. As a result, the study suggested that in the future universities will resort to extension and rebuilding or new building for their facilities. To ensure space efficiency without conflict, we should follow legitimacy of space allocation and composition, building quality university facilities, creating quality environment, preventing tuition from rising or Improving welfare to keep pace with the new era.

• 한국환경복원기술학회지
• /
• 제21권2호
• /
• pp.27-38
• /
• 2018

Environmental planning includes resource allocation and spatial planning process for the conservation and management of environment. Because the spatialization of the environmental planning is not specifically addressed in the relevant statutes, it actually depends on the qualitative methodology such as expert judgement. The results of the qualitative methodology have the advantage that the accumulated knowledge and intuition of the experts can be utilized. However, it is difficult to objectively judge whether it is enough to solve the original problem or whether it is the best of the possible scenarios. Therefore, this study proposed a methodology to quantitatively and objectively spatialize various environmental planning. At first, we suggested a quantitative spatial planning model based on an optimization algorithm. Secondly, we applied this model to two kinds of environmental planning and discussed about the model performance to present the applicability. Since the models were developed based on conceptual study site, there was a limitation in showing possibility of practical use. However, we expected that this study can contribute to the fields related to environmental planning by suggesting flexible and novel methodology.

• Journal of the Korean Data and Information Science Society
• /
• 제28권1호
• /
• pp.11-20
• /
• 2017

시간의 변화뿐만 아니라 공간 위치의 변화를 함께 고려한 자료를 공간 시계열 자료라고 한다. 공간 시계열 자기회귀 이동평균 모형과 공간 시계열 중선형 모형에 대해 소개하고 각각의 Kalman Filter 방법에 의한 모수 추정의 과정을 거쳐 최종 선택된 모형의 예측력을 비교하였다. 또한 공간 시계열 자료의 모형에 포함되는 가중행렬에 대하여 기존의 방법인 동일한 가중치와 더불어 거리에 비례한 가중치와 인구수에 비례한 가중치를 제안하였다. 실증분석을 위해 한국질병관리본부에서 수집한 유행성 이하 선염 자료를 활용하여 가중치를 달리한 공간 시계열 모형을 적합시키고 예측하였다. 예측 오차 제곱합을 활용하여 어느 모형이 가장 효과적인 모형인지 판정하였다.

• Asian Journal of Atmospheric Environment
• /
• 제11권4호
• /
• pp.330-343
• /
• 2017

Climate change is an important issue, with many researches examining not only future climatic conditions, but also the interaction of climate and air quality. In this study, a new version of the emissions processing software tool - Python-based PRocessing Operator for Climate and Emission Scenarios (PROCES) - was developed to support climate and atmospheric chemistry modeling studies. PROCES was designed to cover global and regional scale modeling domains, which correspond to GEOS-Chem and CMAQ/CAMx models, respectively. This tool comprises of one main system and two units of external software. One of the external software units for this processing system was developed using the GIS commercial program, which was used to create spatial allocation profiles as an auxiliary database. The SMOKE-Asia emissions modeling system was linked to the main system as an external software, to create model-ready emissions for regional scale air quality modeling. The main system was coded in Python version 2.7, which includes several functions allowing general emissions processing steps, such as emissions interpolation, spatial allocation and chemical speciation, to create model-ready emissions and auxiliary inputs of SMOKE-Asia, as well as user-friendly functions related to emissions analysis, such as verification and visualization. Due to its flexible software architecture, PROCES can be applied to any pregridded emission data, as well as regional inventories. The application results of our new tool for global and regional (East Asia) scale modeling domain under RCP scenario for the years 1995-2006, 2015-2025, and 2040-2055 was quantitatively in good agreement with the reference data of RCPs.