• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.697-700
• /
• 2020

고위험의 재난 상황에서 사람이 상황을 판단하고, 요구조자를 탐색하며, 구조하는 것은 추가 피해를 발생시킬 수 있다. 따라서 재난 상황에서도 이동과 접근이 용이한 무인항공에 관한 연구와 개발이 활발히 이루어지고 있다. 재난 상황에서 신속하게 대처하기 위해서는 선제적 상황인지 기술이 필요하다. 이에 본 논문은 구조 및 대피를 위해 사람, 자동차, 자전거 등의 객체를 인식하고 중복 인식을 피하기 위해 추적하는 딥러닝 기반 다중 객체 추적 시스템을 제안한다. 2019 인공지능 R&D 그랜드 챌린지 상황인지 부문에서의 대회 결과로 실험 성능을 증명한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.659-662
• /
• 2008

최근 들어 생활 공간과 유비쿼터스 기술 융합의 과정으로 u-City 구현을 통한 상황인식 서비스 제공으로 삶의 질을 향상시키려 노력을 하고 있다. 상황인식 서비스 인프라와 관련된 연구가 활발히 진행되고 있다. u-City는 도시전체를 구성원으로 하고 있기 때문에 센서로부터의 많은 정보의 수집과 다양한 상황으로 다양한 상황정보가 발생할 것이다. 따라서 논문에서는 u-Cit에 적합한 상황인식 추론 시스템에 대해서 제시하고 구현을 통한 제안하는 시스템의 성능에 대해서 논의한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.714-717
• /
• 2008

유비쿼터스 컴퓨팅(Ubiquitous Computing) 시대에 기업의 사무 환경은 다양한 정보들과 많은 사용자들이 유기적인 관계를 형성한다. 이러한 관계에서 접근 제어는 다양한 정보 객체에 허가된 사용자만이 접근할 수 있는 권한을 갖는 기능을 제공하는 것이고, 사무 환경에서 보안상 필수적이며 중요한 역할을 한다. 하지만 기존의 접근 제어 모델들은 상황 정보를 고려하지 않아 동적인 접근 제어를 하지 못하는 문제점을 가지고 있다. 본 논문은 워크플로우 기반의 오피스 환경에서 동적이고 능동적인 접근제어 관리를 제공하기 위한 상황 정보와 역할 기반의 워크플로우 데이터 접근제어 모델을 제안한다. 이 모델은 수많은 상황 정보 및 사무 정보와 사용자가 동적으로 변화하는 사무환경에서 사용자에게 접근을 제어하기 적합하다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.17 no.1
• /
• pp.159-165
• /
• 2017

As increasing the number of the elderly people, various IT-based home systems for the elderly are actively studied. However, while most systems for the elderly focus on systems detecting emergency cases, systems considering energy reduction in the elderly housing are rarely found. In fact, the systems for the elderly housing consumes rather more electric power for their functions. In this paper, a emergency monitoring system including the functions to reduce energy useage is studied and it is implemented. Though the system, we can achieve both of future-oriented functions for the aged and green society, which are emergency monitoring and energy reduction.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.163-170
• /
• 2023

The user-based collaborative filtering technique, one of the implementation methods of the recommendation system, recommends the preferred items of neighboring users based on the calculations of neighboring users with similar rating histories. However, it fundamentally has a data scarcity problem in which the quality of recommendations is significantly reduced when there is little common rating history. To solve this problem, many existing studies have proposed various methods of combining Jaccard index with a similarity measure. In this study, we introduce a time-aware concept to Jaccard index and propose a method of weighting common items with different weights depending on the rating time. As a result of conducting experiments using various performance metrics and time intervals, it is confirmed that the proposed method showed the best performance compared to the original Jaccard index at most metrics, and that the optimal time interval differs depending on the type of performance metric.

• Journal of Intelligence and Information Systems
• /
• v.17 no.1
• /
• pp.1-16
• /
• 2011

The real-time ubiquitous environment is required to be able to process a series of sensor data within limited time. The whole sensor data processing consists of several phases : getting data out of sensor, acquiring context and responding to users. The ubiquitous computing middleware is aware of the context using the input sensor data and a series of data from database or knowledge-base, makes a decision suitable for the context and shows a response according to the decision. When the real-time ubiquitous environment gets a set of sensor data as its input, it needs to be able to estimate the delay-time of the sensor data considering the available resource and the priority of it for scheduling a series of sensor data. Also the sensor data of higher priority can stop the processing of proceeding sensor data. The research field for such a decision making is not yet vibrant. In this paper, we propose a queuing time computation algorithm for sensor data processing in real-time ubiquitous environment.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.1
• /
• pp.34-39
• /
• 2009

As a subjective recognition effect, human's emotion has impulsive characteristic and it expresses intentions and needs unconsciously. These are pregnant with information of the context about the ubiquitous computing environment or intelligent robot systems users. Such indicators which can aware the user's emotion are facial image, voice signal, biological signal spectrum and so on. In this paper, we generate the each result of facial and voice emotion recognition by using facial image and voice for the increasing convenience and efficiency of the emotion recognition. Also, we extract the feature which is the best fit information based on image and sound to upgrade emotion recognition rate and implement Multi-Modal Emotion recognition system based on feature fusion. Eventually, we propose the possibility of the ubiquitous computing service reasoning method based on Bayesian Network and ubiquitous context scenario in the ubiquitous computing environment by using result of emotion recognition.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.7
• /
• pp.942-948
• /
• 2020

In this paper, we propose a clustered algorithm that possible more efficient user distinction within clustering using context-aware attribute information. In typically, the data provided to classify interrelationships within cluster information in the process of clustering data will be as a degrade factor if new or newly processing information is treated as contaminated information in comparative information. In this paper, we have developed a clustering algorithm that can extract user's recognition information to solve this problem in using K-means algorithm. The proposed algorithm analyzes the user's clustering attributed parameters from user clusters using accumulated information and clustering according to their attributes. The results of the simulation with the proposed algorithm showed that the user management system was more adaptable in terms of classifying and maintaining multiple users in clusters.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.5
• /
• pp.755-762
• /
• 2022

This paper propose a clustered algorithm that possible more efficient COVID-19 disease learning prediction within clustering using context-aware attribute information. In typically, clustering of COVID-19 diseases provides to classify interrelationships within disease cluster information in the clustering process. The clustering data will be as a degrade factor if new or newly processing information during treated as contaminated factors in comparative interrelationships information. In this paper, we have shown the solving the problems and developed a clustering algorithm that can extracting disease correlation information in using K-means algorithm. According to their attributes from disease clusters using accumulated information and interrelationships clustering, the proposed algorithm analyzes the disease correlation clustering possible and centering points. The proposed algorithm showed improved adaptability to prediction accuracy of the classification management system in terms of learning as a group of multiple disease attribute information of COVID-19 through the applied simulation results.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.11a
• /
• pp.439-443
• /
• 2007

Nowadays, activity recognition becomes a hot topic in context-aware computing. In activity recognition, machine learning techniques have been widely applied to learn the activity models from labeled activity samples. Most of the existing work uses only one learning method for activity learning and is focused on how to effectively utilize the labeled samples by refining the learning method. However, not much attention has been paid to the use of multiple classifiers for boosting the learning performance. In this paper, we use two methods to generate multiple classifiers. In the first method, the basic learning algorithms for each classifier are the same, while the training data is different (ASTD). In the second method, the basic learning algorithms for each classifier are different, while the training data is the same (ADTS). Experimental results indicate that ADTS can effectively improve activity recognition performance, while ASTD cannot achieve any improvement of the performance. We believe that the classifiers in ADTS are more diverse than those in ASTD.