• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.321-324
• /
• 2003

This paper presents a novel system that analyzes and recognizes a gait based on shape context on silhouette images. The main functions of the system consist of three steps: First, the system extracts the silhouette images from galt image sequence by performing a simple pre-processing and acquires the AGM(Averaged Gait Map) by averaging them. Next. it computes the cross-correlation between the AGMs. Finally, it classifies the AGMs based on the cross-correlation using nearest neighborhood classification. The proposed system uses two cues to classify a gait: One corresponds to biometric shape cue such as body height width. and body-part proportions. The other corresponds to gait cue such as stride length and amount of arm swing. Perceptionally, the biometric cues are sailent on the double support (both legs spread and touching the ground) while the gait cues on the midstance. Through a variety of experiments, it is proved that the property of a gait is mainly influenced by gait cues than biometric cues.

• Proceedings of the KSRS Conference
• /
• 1998.09a
• /
• pp.163-168
• /
• 1998

Extracting a Digital Elevation Model (DEM) from spaceborne imagery is important for cartographic applications of remote sensing data. The procedure for such DEM generation can be divided into stereo matching, sensor modelling and DEM interpolation. Among these, DEM interpolation contributes significantly to the completeness and accuracy of a DEM and, yet, this technique is often considered "trivial". However, na\ulcornere DEM interpolation may result in a less accurate and sometimes meaningless DEM. This paper reports the performance analysis of various DEM interpolation techniques. Using a manually derived DEM as reference, a number of sample points were created randomly. Different interpolation techniques were applied to the sample points to generate DEMs. The performance of interpolation was assessed by the accuracy of such DEMs. The results showed that kriging gave the best results at all times whereas nearest neighborhood interpolation provided a fast solution with moderate accuracy when sample points were large enough.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.109-111
• /
• 2001

기존의 협력적 필터링 기술을 이용한 사용자 선호도 예측 방법에서는 아이템에 대한 사용자의 선호도를 기반으로 이웃 선정 방법(Nearest-Neighborhood Method)을 사용하고, 피어슨 상관 계수에 의해 사용자의 유사도를 구하므로 아이템에 대한 내용을 반영하지 못할 뿐만 아니라 희박성 문제를 해결하지 못하였다. 본 논문에서는 기존의 사용자 선호도 예측 방법의 문제점을 보완하기 위하여 연관 사용자 군집과 베이지안 분류를 이음한 사용자 선호도 예측 방법을 제안한다. 제안한 방법에서는 협력적 필터링 시스템에서의 희박성(Sparsity)문제를 해결하기 위하여 ARHP 알고리즘을 사용하여 사용자를 장르별로 군집하며 새로운 사용자는 Naive Bayes 분류자에 의해 이들 장르 중 하나로 분류된다. 또한, 분류된 장르 내에 속한 사용자들과 새로운 사용자의 유사도출 구하기 위해 Naive Bayes 학습을 통해 사용자가 평가한 아이템에 추정치를 달리 부여한다. 추정치가 부여된 선호도를 기존의 피어슨 상관 관계에 적용할 경우 결측치(Missing Value)로 인한 예측의 오류를 적게 하여 예측의 정확도를 높일 수 있다. 제안된 방법의 성능을 평가하기 위해서 기존의 협력적 필터링 기술과 비교 평가하였다.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.11a
• /
• pp.201-204
• /
• 1997

최근 주소를 자동으로 인식하여 우편물 분류와 같은 업무를 효과적으로 수행하기 위한 연구가 진행되고 있다. 기존 연구들은 낱자 단위의 인식을 수행한 후 사전 형태의 간단한 DB를 통해 최종의 결과를 생성한다. 그러나 한글과 같은 복잡한 구조의 필기 문자에 대한 인식기의 성능은 아직도 미흡한 상태이다. 따라서 낱자 인식기의 성능에 의존하는 현재와 같은 방법으로는 만족할 만한 결과를 얻기가 힘들 것으로 생각된다. 본 논문에서는 낱자 인식 결과에 크게 의존하지 않고 주소에 나타나는 단어의 낱자들 사이간 연결 정보를 이용하여 단어를 인식할 수 있는 시스템을 제안한다. 본 시스템은 통계적 인식기를 사용하여 낱자를 인식하는 부분과 낱자 인식 결과를 조합하여 단어 수준의 인식과정을 통해 최종의 결과를 생성하는 부분으로 구성된다. 통계적 인식기는 Nearest neighborhood 방법을 사용하여 간단한 형태로 구현하였다. 단어인식 모듈은 단어에서 모든 문자간의 관계를 표현할 수 있도록 HMM 모형을 사용하여 어휘정보 네트워크를 구성하고 이를 이용하여 주소에 나타나는 단어를 인식하도록 하였다. PE92 한글 문자 데이터를 이용하여 실험을 수 璿\ulcorner 결과, 통계적 인식기의 성능이 저조함에도 불구하고 HMM을 이용한 어휘정보 네트워크가 이를 보완함으로써 좋은 결과를 얻었다. 이러한 단어 인식 방법을 주소 이외의 다른 단어 집합에 대해서도 쉽게 적용될 수 있을 것으로 예상된다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06b
• /
• pp.220-222
• /
• 2006

현재 마이크로어레이 기술은 대량의 유전자 발현 데이터 특히 암과 관련한 데이터들을 쏟아내고 있다. 이 데이터를 기반으로 암의 종류에 따른 유전자들의 차별적 발현 양상을 분석하고 발현량의 변화가 두드러지는 유전자들에 기반하여 암을 분별할 수 있는 분류 모델을 구축한 후, 이것을 암을 진단하거나 예측하는데 이용할 수 있다. 본 논문에서는 마이크로어레이 데이터를 사용해 특징추출방법과 분류를 위한 Naive Bayes, k-Nearest Neighborhood, Decision Tree, Support Vector Machine, Neural Network 알고리즘을 이용하여 최적의 조합을 찾고 어떤 알고리즘이 가장 효과적인지 실험을 통해 분석해보고 성능평가 하는 것을 목표로 한다.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.11
• /
• pp.1398-1406
• /
• 2004

The nonlinear signals from an impedance meter for the area average void fraction in two-phase flow have been analyzed to construct a phase space trajectory. The pseudo phase space was constructed with the time delay and proper dimensions. The time delay and the embedding dimension were chosen by the average mutual information and by the false nearest neighborhood, respectively. The attractor-density-map of projected states was used to produce the two dimensional probability distribution functions (2D-PDF). Since the developed 2D-PDF showed clear distinction of the flow patterns, the flow regime identification was made with three rules and with the 2D-PDF. Also, the transition criteria of Mishima-Ishii agree well with the present results.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.07a
• /
• pp.95-96
• /
• 2020

본 논문은 발열 검사시에 QR코드를 이용해 1차적인 본인인증 단계 후 K-NN알고리즘을 통한 얼굴인식으로 2차적인 본인인증 을 거친후 비대면식으로 발열검사가 가능한 방법을 제시하였다. 이를 통해서 추적관리 뿐만 아니라 CCTV영상을 통하여 확진자 발생시 인접 인원 추적까지 가능하고, 신속한 추적관리가 가능하게 제공하였다.

• Journal of Environmental Impact Assessment
• /
• v.11 no.1
• /
• pp.51-66
• /
• 2002

The purpose of this study is to analyze the geomorphological environment changes of tidal flat in the Cheonsu Bay. Especially, it centers on the changes in the sedimentary environment using remote sensing data. Multi-temporal Landsat data and topographic maps were used in this study. The results are summarized as follows: the tidal flat of Cheonsu Bay changes in many ways depending on the direction of the tidal current. In the neighborhood of Ganwoldo, the scale of the tidal flat has continuously been expanded due to the superiority of sedimentation after a tide embankment was built. When we analyzed the grain size of sediments and implemented in-situ field survey, it was found that the innermost part of the bay consists of a mud flat, with the midway part mixed flat, and the nearest part to the sea sand flat. On the other hand, in the neighborhood of Seomot isle and its beach, sedimentation is superior in the eastern part whereas erosion is superior in the western part. In other words, the western coast of the beach is contacted with the open seas and under much influence of ocean wave. The eastern coast is placed at the entrance of the bay and has sand bar and tidal flat developed due to submarine deposits that are accumulated on the sea floor by the tidal current. In conclusions, remote sensing methods can be effectively applied for quantitative analysis of geomorphological changes in tidal flat, and it is expected that the proposed schemes can be applied to another geomorphological environments such as beach, sand dune, and sand wave.

• Journal of Cadastre & Land InformatiX
• /
• v.48 no.2
• /
• pp.185-195
• /
• 2018

In the Seoul Metropolitan City, the PM(pariculate matter) application used by the citizens provides the PM concentration of the nearest monitoring stations located on the PM monitoring stations. Currently, the selecting method of the PM monitoring network considered by the Ministry of Environment is based on considering the monitoring station distribution and population density only. In this study, we analyzed the distance between PM monitoring station and the administrative center point in addition to the above considerations. The number of test sites was verified and the range of coverage of each monitoring stations was indicated by using the Voronoi algorithm and hexagon grid. The spatial position problem of the PM monitoring station was suggested by spatial data analysis. The variables of spatial data analysis are single-family houses, apartments, $1^{st}$ class neighborhood, $2^{nd}$ class neighborhood, garbage disposal plant, hazardous material disposal facility, factory, and the density map. The analysis result of the selection criterion considering the additional variables for new PM monitoring stations was presented, in addition to the selection criteria provided by the Ministry of Environment.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.10 no.2
• /
• pp.164-171
• /
• 2007

This research used GIS spatial analysis model and Huff probability model and achieved trade area analysis of area center. we constructed basic maps that were surveyed according to types of business, number of households etc. using a land registration map of LMIS(Land Management Information System) in Bokdae-dong, Cheongju-si. Kernel density function and NNI(Nearest Neighbor Index) was used to estimate store distribution center area in neighborhood life zones. The center point of area and scale were estimated by means of the center area. Huff probability model was used in abstracting trade areas according to estimated center areas, those was drew map. Therefore, this study describes method that can apply in Huff probability model through kernel density function and NNI of GIS spatial analysis techniques. A trade area was abstracted more exactly by taking advantage of this method, which will can aid merchant for the foundation of small sized enterprises.