• Annual Conference of KIPS
• /
• 2007.05a
• /
• pp.305-308
• /
• 2007

본 논문은 교통표지판을 자동으로 인식하는 방법에 관한 연구로 기존의 교통표지판 인식시스템에서는 인식하는데 걸리는 시간이 길고 잡음환경에서 인식률이 저하되며 변경된 교통표지판은 인식하지 못하는 문제점이 있다. 본 논문에서는 이와 같은 문제점을 해결하기위해 컬러정보를 이용하여 교통표지판 영역을 추출하고 추출된 이미지를 인식하는데 다층퍼셉트론 신경망 알고리즘을 적용하여 교통표지판 인식시스템을 제안한다. 제안된 방법은 교통표지판의 컬러를 분석하여 영상에서 교통표지판 영역을 추출한다. 영역을 추출하는 방법은 RGB 컬러 공간으로부터 YUV, YIQ, CMYK 컬러 공간이 가지는 특성을 이용한다. 형태처리는 교통표지판의 기하학적 특성을 이용하여 군집화한다. 교통표지판 인식은 학습이 가능한 다층퍼셉트론의 오류역전파알고리즘을 적용하여 인식한다. 다층퍼셉트론 신경망 알고리즘은 패턴인식 분야에서 우수한 성능이 입증 되었다.

• Journal of the Korea Computer Graphics Society
• /
• v.10 no.2
• /
• pp.42-50
• /
• 2004

캐릭터 애니메이션 기술의 발달로 가상공간에 애니메이트되는 캐릭터의 수가 점점 증가되고 있으며, 캐릭터 자체 골격구조의 관절 개수와 캐릭터를 덮고 있는 메쉬의 폴리곤 개수도 점점 증가하는 추세이다. 따라서, 실시간 가상환경에서 다수의 캐릭터를 전처리 과정 없이 시뮬레이션할 경우 전체 군중시스템 성능의 저하가 예상된다. 본 논문에서는, 이러한 문제점을 해결하기 위해 모션 다단계(motion level-of-detail) 기법을 제시한다. 모션 단순화 기법은 캐릭터의 움직임을 제어하는 골격(관절)구조와 캐릭터의 형태를 시각적으로 표현하는 기하(메쉬)구조를 단순화 하는 방법으로 기존 동작과 단순화된 동작의 차이를 최소화 한다. 골격구조 단순화를 위한 JPC(joint posture clustering)방법은 특정 관절의 연속된 모션 시퀀스에서의 유사 자세 집단을 추출하여 하나의 자세로 표현하는 방법으로, 모션의 특성에 따라 동적으로 관절을 단순화하여 관절 시뮬레이션 시간을 줄이는 방법이다. JPC방법은 골격구조가 시간에 따라 동적으로 변형되기 때문에 골격구조의 계층구조를 재 구축할 시간이 필요하지만, 기존 동작과 유사성을 잃지 않는 단순화된 동작 생성이 가능하다. 유사 자세 집단을 추출하기 위해 전체 모션 시퀀스에서 관절의 프레임간 자세 차이를 수식화하여 테이블 형태로 구성하고 이를 통해 기존 동작의 유사성을 잃지 않으며 관절의 단순화 율을 최대화 할 수 있는 알고리즘을 제시한다. 또한, 실시간 군중 환경의 성능을 더욱 향상시키기 위해 시간에 따라 변형되는 캐릭터 메쉬의 단순화 기법을 적용한다. 실험결과 모션 다단계 기법은 실시간 군중환경에서 캐릭터의 수가 많고 복잡한 골격구조와 기하구조로 구성된 관절 궤적의 변화가 심하지 않은 동작에 대해 특히 효율적이다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.2
• /
• pp.145-150
• /
• 2015

Needs about safety of residents are important in urbanized society, elderly and small-size family. People are looking for safety information system and device of CPTED. That is, Needs and Installations of CCTV increased steadily. But, scientific analysis about validity, systematic plan and location of security CCTV is nonexistent. It is simply put these devised in more demanded areas. It has limits to look for safety of residents by increasing density of CCTVs. One of the characteristics of crime is clustering and stong interconnectivity. So, exploratory spatial data of crime is geo-coded using 2 years data and carried out cluster analysis and space statistical analysis through GIS space analysis by dividing 18 variables into social economy, urban space, crime prevention facility and crime occurrence index. The result of analysis shows cluster of 5 major crimes, theft, violence and sexual violence by Nearest Neighbor distance analysis and Ripley's K function. It also shows strong crime interconnectivity through criminal correlation analysis. In case of finding criminal cluster, you can find criminal hotspot. So, in this study I found concept of hotspot and considered technique about selection of hotspot. And then, selected hotspot about 5 major crimes, theft, violence and sexual violence through Nearest Neighbor Hierarchical Spatial Clustering.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.15 no.4
• /
• pp.162-174
• /
• 2012

To visualize the characteristics of vertical seawater temperature data, in the ocean having 3D spatial characteristics, 2D thematic maps like horizontal seawater temperature distribution map at each depth layer and 3D volume model using 3D spatial interpolation are used. Although these methods are useful to understand oceanographic phenomena visually, there is a limit to analyze the spatial pattern of vertical temperature distribution or the relationship between vertical temperature characteristics and other oceanic factors (seawater chemistry, marine organism, climate change, etc). Therefore, this study aims to determine the spatial distribution characteristics of vertical temperature profiles in the South Sea of Jeju by quantifying the characteristics of vertical temperature profiles by using an algorithm that can extract the thermocline parameters, such as mixed layer depth, maximum temperature gradient and thermocline thickness. For this purpose spatial autocorrelation index (Moran's I) was calculated including mapping of spatial distribution for three parameters representing the vertical temperature profiles. Also, after grouping study area as four regions by using cluster analysis with three parameters, the characteristics of vertical temperature profiles were defined for each region.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.4
• /
• pp.391-396
• /
• 2003

This paper proposes a new approach that converts continuous-valued attributes to categorical-valued ones considering the distribution of target attributes(classes). In this approach, It can be possible to get optimal interval boundaries by considering the distribution of data itself without any requirements of parameters. For each attributes, the distribution of target attributes is projected to one-dimensional space. And this space is clustered according to the criteria like as the density value of each target attributes and the amount of overlapped areas among each density values of target attributes. Clusters which are made in this ways are based on the probabilities that can predict a target attribute of instances. Therefore it has an interval boundaries that minimize a loss of information of original data. An improved performance of proposed discretization method can be validated using C4.5 algorithm and UCI Machine Learning Data Repository data sets.

• Journal of the Korean association of regional geographers
• /
• v.15 no.4
• /
• pp.510-523
• /
• 2009

Geodemographics(GD) can be defined as an analytical approach of socio-economic and behavioral data about people to investigate geographical patterns. GD is based on the assumptions that demographical and behavioral characteristics of people who live in the same neighborhood are similar and then the neighborhoods can be categorized with spatial classifications with the geographical classifications. Thus, this paper, in order to identify the applicability of the geographical classification of the GD, explores the concepts of the geodemographics into Seoul city areas with Korea census data sets that contain key characteristics of demographic profiles in the area. Then, this paper attempt to explain each area classification profile by using clustering techniques with Ward's and k-means statistical methods. For this as as as, this paper employs 2005 Census dataset released by Korea National Statistics Office and the neighborhood unit is based on Dong level, the smallest administrative boundary unit in Korea. After selecting and standardizing variables, several areas are categorized by the cluster techniques into 13, this paps as distinctive cluster profiles. These cluster profiles are used to cthite a short description and expand on the cluster names. Finally, the results of the classification propose a reasonable judgement for target area types which benefits for the people who make a spatial decision for their spatial problem-solving.

• Journal of Platform Technology
• /
• v.11 no.5
• /
• pp.23-36
• /
• 2023

This study aims to analyze the similarities and dissimilarities of various versions of <Heungbu-jeon> at both micro- and macro-levels using contents analysis techniques and the Hamming distance metrics. The 28 versions of <Heungbu-jeon> were segmented into 341 content units, and for each unit, the value of the content type was encoded. The dissimilarities between content types were compared among all versions by the content unit, respectively. The (dis-)similarities based on the content types of the 28 versions were aggregated and transformed into a distance matrix. The matrix was interpreted by multi-dimensional scaling, resulting into the two-dimensional coordinates. By visualizing the results by multi-dimensional scaling analysis, it was confirmed that the versions of <Heungbu-jeon> can be broadly divided into two groups. Hierarchical clustering and phylogenetic analysis were applied to analyze the clusters of the 28 versions, using the same distance matrix. The results showed that there are five clusters based on the micro-level analysis of (dis-)similarities within two major clusters. This study demonstrated the usefulness of applying digital humanities methods to encode the content of classical literary versions and analyze the data using clustering analysis techniques based on the (dis-)similarity of literary content.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.3
• /
• pp.37-46
• /
• 2013

Similarity measure of catchments is essential for regionalization studies, which provide in depth analysis in hydrological response and flood estimations at ungauged catchments. However, this similarity measure is often biased to the selected catchments and is not clearly explained in hydrological sense. This study applied a type of hydrological similarity distance measure-Flood Estimation Handbook to 25 Geum River catchments, Korea. Three Catchment Characteristics, Area(A)-Annual precipitation(SAAR)-SCS Curve Number(CN), are used in Euclidian distance measures. Furthermore, six index of Flow Duration Curve are applied to clustering analysis of SPSS. The catchments' grouping of hydrological similarity measures suggests three groups (H1, H2 and H3) and the four catchments are not grouped in this study. The clustering analysis of FDC provides four Groups; F1, F2, F3 and F4. The six catchments (out of seven) of H1 are grouped in F1, while Sangyeogyo is grouped in F2. The four catchments (out of six) of H2 are also grouped in F2, while Cheongju and Guryong are grouped in F1. The catchments of H3 are categorized in F1. The authors examine the results (H1, H2 and H3) of similarity measure based on catchment physical descriptors with results (F1 and F2) of clustering based on catchment hydrological response. The results of hydrological similarity measures are supported by clustering analysis of FDC. This study shows a potential of hydrological catchment similarity measures in Korea.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.4
• /
• pp.353-362
• /
• 2012

3D building modeling from airborne Lidar without model regularization may cause positional errors or topological inconsistency in building models. Regularization of 3D building models, on the other hand, restricts the types of models which can be reconstructed. To resolve these issues, this paper modelled 3D buildings from airborne Lidar by building model regularization which considers more various types of buildings. Building points are first segmented into roof planes by clustering in feature space and segmentation in object space. Then, 3D building models are reconstructed by consecutive adjustment of planes, lines, and points to satisfy parallelism, symmetry, and consistency between model components. The experimental results demonstrated that the method could make more various types of 3d building models with regularity. The effects of regularization on the positional accuracies of models were also analyzed quantitatively.

• The Journal of the Korea Contents Association
• /
• v.4 no.3
• /
• pp.81-90
• /
• 2004

Recently, It has been increased to use a multi-dimensional data in various applications with a rapid growth of the computing environment. In this paper, we propose the vector approximate tree for content-based retrieval of multi-dimensional data. The proposed index structure reduces the depth of tree by storing the many region information in a node because of representing region information using space partition based method and vector approximation method. Also it efficiently handles 'dimensionality curse' that causes a problem of multi-dimensional index structure by assigning the multi-dimensional data space to dynamic bit. And it provides the more correct regions by representing the child region information as the parent region information relatively. We show that our index structure outperforms the existing index structure by various experimental evaluations.