• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.114-117
• /
• 2003

본 논문에서는 video stream내의 움직이는 객체 정보를 추정하고 동적 GTM(genetic tree-map) 알고리즘을 사용하여 얼굴 영역 검출 기법을 제안한다. 기존의 일반적인 객체 추정 기법은 클러스터(cluster)과정을 통하여 영상 정보를 분할하고 그 중 움직이는 객체 부분을 복원함으로서 추정하였다. 제안하는 기법은 BMA(block matching algorithm)［1］ 알고리즘을 사용하여 video stream 에서 움직이는 객체 정보를 얻고 클러스터 알고리즘으로 PCA(principal component analysis)를 사용한다. PCA 기법은 입력 데이터에 관해 통계적 특성을 이용하여 주성분을 찾는다. 주축과 영역분할 알고리즘을 사용하여 데이터를 분할하고, 분할된 객체 정보를 사용하여 특정 객체만을 추정하는 것이 가능하다. 이렇게 추정된 객체를 얼굴영역의 feature에 대하여 신경망 학습인 동적 GTM 알고리즘을 사용하여 생성된 동적 GTM 맵의 정보에 따라 객체의 얼굴영역만을 추출해 낼 수 있다［2-6］.

• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.430-432
• /
• 1993

In this paper, an algorithm for the fault diagnosis using simple FCM(Fuzzy Cognitive Map) is proposed FCMs which store uncertain causal knowledges are fuzzy signed graphs with feedback. The algorithm allows searching the origin of fault and the ways of propagating the abnormality throughout the process simply and has following characteristics. First, it can distinguish the cause of soft failure which can degenerate the process as well as hard failure. Second, it is proper for the processes which have difficulties to establish the exact quantative model. Finally, it has short amputation time in comparison with the fault tree or the other AI methods. The applicability of the proposed algorithm for the fault diagonosis to a tank or pipeline system is demonstrated

• Journal of the Korea Institute of Military Science and Technology
• /
• v.16 no.6
• /
• pp.726-732
• /
• 2013

It is required for UGV(Unmanned Ground Vehicle) to have a LPP(Local Path Plan) component which generate a local path via the center of road by analyzing binary map to travel autonomously unpaved road in rough environment. In this paper, we present the method of boundary estimation for unpaved road and a local path planning method based on RANGER algorithm using the estimated boundary. In specially, the paper presents an approach to estimate road boundary and the selection method of candidate path to minimize the problem of zigzag driving based on Bayesian probability reasoning. Field test is conducted with scenarios in rough environment in which bush, tree and unpaved road are included and the performance of proposed method is validated.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.529-531
• /
• 2022

기후 위기가 대두되며 탄소중립에 많은 관심이 쏟아지고 있다. 탄소중립을 실천하기 위한 여러 가지 방법 중 도시의 수목을 관리하는 것은 탄소배출 저감, 대기질 개선 등의 환경적인 긍정적 효과를 얻을 수 있다. 수종별 온실가스 흡수량과 흡수 계수에는 차이가 있지만 도시 나무 캐노피를 증가시키면 온실가스 흡수량도 증가한다. 본 논문은 탄소정보공개 프로젝트(CDP)에서 제공하는 데이터를 기반으로 도시의 녹지 지대를 구글 지도(Google Map) 위성사진을 통해 찾아내고 지니 계수(Gini Coefficient)를 통해 도심 녹지 균형을 비교하였다. 향후 도시 수목과 녹지 데이터를 축적해 기초자료가 쌓이면 도시환경의 지표로 활용될 수 있을 것으로 기대된다.

• Journal of Cadastre & Land InformatiX
• /
• v.46 no.2
• /
• pp.27-42
• /
• 2016

Forest accounts for almost 64 percents of total land cover in South Korea. For inventorying, monitoring, and managing such large area of forest, application of remote sensing and geographic information system (RS/GIS) technology is essential. On the basis of spectral characteristics of satellite imagery, forest cover and tree species can be classified, and forest cover map can be prepared. Using three dimensional data of LiDAR(Light Detection and Ranging), tree location and tree height can be measured, and biomass and carbon stocks can be also estimated. In addition, many indices can be extracted using reflection characteristics of land cover. For example, the level of vegetation vitality and forest degradation can be analyzed with VI (vegetation Index) and TGSI (Top Grain Soil Index), respectively. Also, pine wilt disease and o ak w ilt d isease c an b e e arly detected and controled through understanding of change in vegetation indices. RS and GIS take an important role in assessing carbon storage in climate change related projects such as A/R CDM, REDD+ as well. In the field of climate change adaptation, impact and vulnerability can be spatio-temporally assessed for national and local level with the help of spatio-temporal data of GIS. Forest growth, tree mortality, land slide, forest fire can be spatio-temporally estimated using the models in which spatio-temporal data of GIS are added as influence variables.

• Journal of Korean Society of Forest Science
• /
• v.62 no.1
• /
• pp.1-18
• /
• 1983

Sum of daily mean temperature, sum of daily mean relative humidity and sum of daily mean duration of sunshine during the growing season (March-October), and daily mean temperature, daily mean relative humidity and daily mean minimum temperature during the dormant season (November-February) were obtained respectively from the climactic data recorded at 26 different standard stations for 30 years from 1951 to 1980, to provide a method for proper selection of tree species suitable to a certain site. They were also marked on the map of Korea. The whole country was divided into 6 regions by trend of temperature variation and the regression equations for each region were produced to estimate the sum of daily mean temperature of the growing season and the sum of daily mean minimum temperature of the dormant season in a certain site where tree plantings are planned. The natural range of distribution of each species was expressed by the sum of daily mean temperature and daily mean minimum temperature on the basis of "Horizontal and vertical distribution of the Korean woody plants" reported by Chung and Lee (1965).

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.5
• /
• pp.65-73
• /
• 2017

In this paper, we propose a stereo matching method to improve the image quality at the hole and the disparity discontinuity regions. The stereo matching method extracts disparity map finding corresponding points between stereo image pair. However conventional stereo matching methods have a problem about the tradeoff between accuracy and precision with respect to the length of the baseline of the stereo image pair. In addition, there are hole and disparity discontinuity regions which are caused by textureless regions and occlusion regions of the stereo image pair. The proposed method extracts initial disparity map improved at disparity discontinuity and miss-matched regions using modified AD-Census-Gradient method and adaptive weighted cost aggregation. And then we conduct the disparity map refinement to improve at miss-matched regions, while also improving the accuracy of the image. Experimental results demonstrate that the proposed method produces high-quality disparity maps by successfully improving miss-matching regions and accuracy while maintaining matching performance compared to existing methods which produce disparity maps with high matching performance. And the matching performance is increased about 3.22(%) compared to latest stereo matching methods in case of test images which have high error ratio.

• Journal of KIISE:Databases
• /
• v.35 no.5
• /
• pp.447-458
• /
• 2008

In this paper, we address an efficient processing scheme for k-nearest neighbor queries to retrieve k static objects in road network databases. Existing methods cannot expect a query processing speed-up by index structures in road network databases, since it is impossible to build an index by the network distance, which cannot meet the triangular inequality requirement, essential for index creation, but only possible in a totally ordered set. Thus, these previous methods suffer from a serious performance degradation in query processing. Another method using pre-computed network distances also suffers from a serious storage overhead to maintain a huge amount of pre-computed network distances. To solve these performance and storage problems at the same time, this paper proposes a novel approach that creates an index for moving objects by approximating their network distances and efficiently processes k-nearest neighbor queries by means of the approximate index. For this approach, we proposed a systematic way of mapping each moving object on a road network into the corresponding absolute position in the m-dimensional space. To meet the triangular inequality this paper proposes a new notion of average network distance, and uses FastMap to map moving objects to their corresponding points in the m-dimensional space. After then, we present an approximate indexing algorithm to build an R*-tree, a multidimensional index, on the m-dimensional points of moving objects. The proposed scheme presents a query processing algorithm capable of efficiently evaluating k-nearest neighbor queries by finding k-nearest points (i.e., k-nearest moving objects) from the m-dimensional index. Finally, a variety of extensive experiments verifies the performance enhancement of the proposed approach by performing especially for the real-life road network databases.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.7 no.2
• /
• pp.132-140
• /
• 2005

This study was conducted to develop regional site index equations for main tree species in Gangwon, Gyunggi-Chungcheong, Gyungsang, and Jeolla area of Korea, using environmental and soil factors obtained from a digital forest site map. Using the large data set obtained from the digital forest map, a total of 28 environmental and soil factors were regressed on site index by tree species for developing the best site index equations for each of the regions. The selected main tree species were Larix 1eptolepis, Pinus koraiensis, Pinus densiflora, Pinus thunbergii, and Quercus acutissima. Finally, four to five environmental and soil factors by species were chosen as independent variables in defining the best regional site index equations with the highest coefficients of determination $(R^2)$. For those site index equations, three evaluation statistics such as mean difference, standard deviation of difference and standard error of difference were applied to the data sets independently collected from fields within the region. According to the evaluation statistics, it was found that the regional site index equations by species developed in this study conformed well to the independent data set, having relatively low bias and variation. It was concluded that the regional site index equations by species had sufficient capability for the estimation of site productivity.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.21 no.4
• /
• pp.64-80
• /
• 2018

In Korea, citizens can only know general information about crime. Thus it is difficult to know how much they are exposed to crime. If the police can predict the crime risky area, it will be possible to cope with the crime efficiently even though insufficient police and enforcement resources. However, there is no prediction system in Korea and the related researches are very much poor. From these backgrounds, the final goal of this study is to develop an automated crime prediction system. However, for the first step, we build a big data set which consists of local real crime information and urban physical or non-physical data. Then, we developed a crime prediction model through machine learning method. Finally, we assumed several possible scenarios and calculated the probability of crime and visualized the results in a map so as to increase the people's understanding. Among the factors affecting the crime occurrence revealed in previous and case studies, data was processed in the form of a big data for machine learning: real crime information, weather information (temperature, rainfall, wind speed, humidity, sunshine, insolation, snowfall, cloud cover) and local information (average building coverage, average floor area ratio, average building height, number of buildings, average appraised land value, average area of residential building, average number of ground floor). Among the supervised machine learning algorithms, the decision tree model, the random forest model, and the SVM model, which are known to be powerful and accurate in various fields were utilized to construct crime prevention model. As a result, decision tree model with the lowest RMSE was selected as an optimal prediction model. Based on this model, several scenarios were set for theft and violence cases which are the most frequent in the case city J, and the probability of crime was estimated by $250{\times}250m$ grid. As a result, we could find that the high crime risky area is occurring in three patterns in case city J. The probability of crime was divided into three classes and visualized in map by $250{\times}250m$ grid. Finally, we could develop a crime prediction model using machine learning algorithm and visualized the crime risky areas in a map which can recalculate the model and visualize the result simultaneously as time and urban conditions change.