• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.8
• /
• pp.3965-3983
• /
• 2017

We propose a novel eye tracking method that can estimate 3D world coordinates using an infrared (IR) stereo camera for indoor and outdoor environments. This method first detects dark evidences such as eyes, eyebrows and mouths by fast multi-level thresholding. Among these evidences, eye pair evidences are detected by evidential reasoning and geometrical rules. For robust accuracy, two classifiers based on multiple layer perceptron (MLP) using gradient local binary patterns (GLBPs) verify whether the detected evidences are real eye pairs or not. Finally, the 3D world coordinates of detected eyes are calculated by region-based stereo matching. Compared with other eye detection methods, the proposed method can detect the eyes of people wearing sunglasses due to the use of the IR spectrum. Especially, when people are in dark environments such as driving at nighttime, driving in an indoor carpark, or passing through a tunnel, human eyes can be robustly detected because we use active IR illuminators. In the experimental results, it is shown that the proposed method can detect eye pairs with high performance in real-time under variable illumination conditions. Therefore, the proposed method can contribute to human-computer interactions (HCIs) and intelligent transportation systems (ITSs) applications such as gaze tracking, windshield head-up display and drowsiness detection.

• The Journal of Korea Robotics Society
• /
• v.8 no.4
• /
• pp.256-265
• /
• 2013

A local map-based exploration algorithm for mobile robots is presented. Segmented frontiers and their relative transformations constitute a tree structure. By the proposed efficient frontier segmentation and a local map management method, a robot can reduce the unknown area and update the local grid map which is assigned to each frontier node. Although this local map-based exploration method uses only local maps and their adjacent node information, mapping completion and efficiency can be greatly improved by merging and updating the frontier nodes. Also, we suggest appropriate graph search exploration methods for corridor and hall environments. The simulation demonstrates that the entire environment can be represented by well-distributed frontier nodes.

• International Journal of Control, Automation, and Systems
• /
• v.5 no.5
• /
• pp.577-583
• /
• 2007

Simultaneous Localization and Mapping (SLAM) is the process of building a map of an unknown environment and simultaneously localizing a robot relative to this map. SLAM is very important for the indoor navigation of a mobile robot and much research has been conducted on this subject. Although feature-based SLAM using an Extended Kalman Filter (EKF) is widely used, it has shortcomings in that the computational complexity grows in proportion to the square of the number of features. This prohibits EKF-SLAM from operating in real time and makes it unfeasible in large environments where many features exist. This paper presents an algorithm which reduces the computational complexity of EKF-SLAM by using topological information (TI) extracted through a thinning process. The global map can be divided into local areas using the nodes of a thinning-based topological map. SLAM is then performed in local instead of global areas. Experimental results for various environments show that the performance and efficiency of the proposed EKF-SLAM/TI scheme are excellent.

• Journal of Institute of Control, Robotics and Systems
• /
• v.2 no.3
• /
• pp.200-208
• /
• 1996

This paper describes the mechanism, guidance, sensor system, and navigation algorithm of METRO, a free ranging mobile robot. METRO is designed for use in structured surroundings or factory environments rather than unstructured natural environments. An overview of the physical configuration of the mobile robot is presented as well as a description of its sensor system, an omnidirectional laser range finder. Except for the global path planning algorithm, a guidance and a navigation algorithm with a local path planning algorithm are used to navigate the mobile robot. In METRO the computer support is divided into a supervisor with image processing and local path planning and a slave with motor control. The free ranging mobile robot is self-controlled and all processing being performed on board.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.5
• /
• pp.1303-1311
• /
• 1999

In order to learn in dynamic environments, reinforcement learning algorithms like Q-learning, TD(0)-learning, TD(λ)-learning have been proposed. however, most of them have a drawback of very slow learning because the reinforcement value is given when they reach their goal state. In this thesis, we have proposed a reinforcement learning method that can approximate fast to the goal state in maze environments. The proposed reinforcement learning method is separated into global learning and local learning, and then it executes learning. Global learning is a learning that uses the replacing eligibility trace method to search the goal state. In local learning, it propagates the goal state value that has been searched through global learning to neighboring sates, and then searches goal state in neighboring states. we can show through experiments that the reinforcement learning method proposed in this thesis can find out an optimal solution faster than other reinforcement learning methods like Q-learning, TD(o)learning and TD(λ)-learning.

• Journal of Information Processing Systems
• /
• v.3 no.1
• /
• pp.8-15
• /
• 2007

Grid environments provide the mechanism to share heterogeneous resources among nodes. Because of the similarity between grid environments and P2P networks, the structures of P2P networks can be adapted to enhance scalability and efficiency in deployment and to search for services. In this paper, we present a membership management based on a hierarchical ring which constructs P2P-like Grid environments. The proposed approach uses only a limited number of connections, reducing communication cost. Also, it only keeps local information for membership, which leads to a further reduction in management cost. This paper analyzes the performance of the approach by simulation and compares it with other approaches.

• Journal of Korea Multimedia Society
• /
• v.17 no.3
• /
• pp.353-363
• /
• 2014

Cloud computing is an internet-based technology, providing services to the virtualized IT environment, and allowing users to add or remove resources of hardware or software at their discretion. Since Cloud computing can construct virtually integrated environments out of multiple local computing environments, various information services can be provided by it. In addition, state organizations also strive to build the cloud computing environments due to the benefits of reduced costs to introduce the system and of reduced time to build and provide the IT services. This study suggests a web-based cloud computing system for the computing environments, to be applied for the Korean National Health and Nutrition Examination Survey (KNHANES) by the Ministry of Health and Welfare, Republic of Korea.

• Journal of the Korean housing association
• /
• v.19 no.6
• /
• pp.73-83
• /
• 2008

We explore what factors influence the residential choice and mobility of people, and try to figure out the preferable residential district in Deagu. Household, housing characteristics, and neighborhood environments are considered as explanatory variables to predict the residential choice and the preferable residential district, and logit regression is used for the analysis. We found age, ownership, income, property, and education level as household characteristics, building age as housing characteristics, and accessibility to park and open space, public library and shopping mall as neighborhood environments are significant in determining residential choice of people whereas housing size, accessibility to elementary school, local market, cultural facility and gymnasium are not significant. These results imply people choose the residential district according to household characteristics as they did, as well as choose according to housing characteristics and neighborhood environments which are newly issued factors.

• Steel and Composite Structures
• /
• v.24 no.6
• /
• pp.697-709
• /
• 2017

The current study addresses the local buckling analysis of an infinite thin rectangular symmetrically laminated composite plate restrained by a tensionless Winkler foundation and subjected to uniform in-plane shear loading. An analytic method (i.e., one-dimensional mathematical method) is used to achieve the analytical solution estimate of the contact buckling coefficient. In addition, to study the effect of ply angle and foundation stiffness on the critical buckling coefficients for the laminated composite plates, the parametric studies are implemented. Moreover, the convergence for finite element (FE) mesh is analysed, and then the examples in the parametric study are validated by the FE analysis. The results show that the FE analysis has a good agreement with the analytical solutions. Finally, an example with the analytical solution and FE analysis is presented to demonstrate the availability and feasibility of the presented analytical method.

• Journal of Korean Society for Atmospheric Environment
• /
• v.18 no.4
• /
• pp.253-264
• /
• 2002

Surface ozone concentrations measured at 40 monitoring sites in three major cities (Seoul, Busan, and Daegu) of Korea during 1993~2000 were analyzed to understand the characteristics of temporal and spatial distributions. Trends were analyzed for annual mean, 95th percentiles of daily 8-hour maximum and days exceeding 8-h ozone standard of 60 ppb. Three indicators exhibited increasing trends (+0.75 ppb yr$^{-1}$ , +2.20 ppb yr$_{-1}$ , and +5.35 days yr$_{-1}$ on average) throughout the study period at all cities. Diurnal and seasonal variations were the largest in Seoul followed by Daegue and Busan, due to the high photochemical production and titration of ozone (Seoul), strong wind and constant supply of background ozone from the ocean (Busan). In the urban centers and industrial areas at all cities, scavenging of ozone by NO reduces the daily 8-hour maximum ozone by 10 ppb on average. High concentrations of ozone have frequently occurred in downwind eastern (Seoul and Daegu) or northern (Busan) sides of the territory. In particular, the coastal area of Busan had relatively high ozone level due to the local sea land breeze circulation. The results indicated that the temporal and spatial variations of ozone concentration were non -uniform and were closely related to the local environments; emission levels, climates, and geographic locations.