• Journal of KIISE:Software and Applications
• /
• v.26 no.1
• /
• pp.157-157
• /
• 1999

Researches on the gesture recognition have become a very interesting topic in the computer vision area, Gesture recognition from visual images has a number of potential applicationssuch as HCI (Human Computer Interaction), VR(Virtual Reality), machine vision. To overcome thetechnical barriers in visual processing, conventional approaches have employed cumbersome devicessuch as datagloves or color marked gloves. In this research, we capture gesture images without usingexternal devices and generate a gesture trajectery composed of point-tokens. The trajectory Is spottedusing phase-based velocity constraints and recognized using the discrete left-right HMM. Inputvectors to the HMM are obtained by using the LBG clustering algorithm on a polar-coordinate spacewhere point-tokens on the Cartesian space .are converted. A gesture vocabulary is composed oftwenty-two dynamic hand gestures for editing drawing elements. In our experiment, one hundred dataper gesture are collected from twenty persons, Fifty data are used for training and another fifty datafor recognition experiment. The recognition result shows about 95% recognition rate and also thepossibility that these results can be applied to several potential systems operated by gestures. Thedeveloped system is running in real time for editing basic graphic primitives in the hardwareenvironments of a Pentium-pro (200 MHz), a Matrox Meteor graphic board and a CCD camera, anda Window95 and Visual C++ software environment.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.11a
• /
• pp.146-149
• /
• 2014

최근 자동차 산업과 IT 기술의 융합이 활발해지면서 스마트카, 자율주행 자동차(무인 자동차)와 같은 지능형 자동차 개발이 활발히 진행되고 지능형 자동차의 비전 기반 기술개발도 활발히 진행되고 있다. 고속도로와 같이 포장된 도로나 자갈길과 같은 비포장 도로에서도 운전자의 승차감을 고려한 능동적 안전시스템과 안정적인 자율주행 자동차의 주행능력을 보장하는 기술들 중 도로 유형을 판단하는 것이 중요 요소 중 하나이다. 따라서 본 논문에서는 가중치 기반 클러스터링 기술을 이용하여 도로표면 유형을 분류하는 알고리즘을 제안한다. 아스팔트, 자갈길, 흙길, 눈길의 도로표면 영상 데이터를 히스토그램의 분포도와 최고점 위치, 에지 영상의 에지량, 채도성분을 이용하여 특징값을 추출하고 클러스터를 구성한다. 분류할 입력 도로표면 영상에 대해 특징값을 분석한 후 탐색범위 내 선택된 각 클러스터의 벡터와의 거리를 측정하여 가중치를 계산하고 가중치가 높은 클러스터를 분류하여 입력 영상에 대한 도로표면을 결정한다. 실험결과 제안하는 방법이 각 도로표면 영상의 특징값과 이를 이용한 가중치만을 이용하여 약 91.25%의 정확도로 도로의 표면을 분류해 내는 것을 볼 수 있었다.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.9
• /
• pp.2350-2364
• /
• 1999

In this paper, we propose a method of class library repository design for provide reuser the object-oriented C++ class component. To class library design, we started by studying the characteristics of a reusable component. We formally defined the reusable component model using an entity relationship model. This formal definition has been directly used as the database schema for storing the reusable component in a repository. The reusable class library may be considered a knowledge base for software reuse. Thus, we used that Enumerative classification of breakdown of knowledge based. And another used classification is clustering of based on class similarity. The class similarity composes member function similarity and member data similarity. Finally, we have designed class library for hierarchical inheritance mechanism of object-oriented concept Generalization, Specialization and Aggregation.

• Proceedings of the Korea Contents Association Conference
• /
• 2008.05a
• /
• pp.547-550
• /
• 2008

Recently, changes in the environment of user-centric internet service such as blog, UCC and IPTV and ubiquitous computing based on web service are needed to high availability system platform. High availability system is to provide safe service continuously even if system failure occurs in clustering system at the network. And it is necessary to synchronize data for reliable service in high availability system. In this paper, I implement DRBD(Disk Replicated Block Device) which is synchronization technique for data of local disks in high availability system.

• Journal of Internet Computing and Services
• /
• v.17 no.6
• /
• pp.153-158
• /
• 2016

As the number of social network users increases, the information on event such as social issues and disasters receiving attention in each region is promptly posted by the bucket through social media site in real time, and its social ripple effect becomes huge. This study proposes a detection method of events that draw attention from users in specific region at specific time by using twitter data with regional information. In order to collect Twitter data, we use Twitter Streaming API. After collecting data, We implemented event detection system by analyze the frequency of a keyword which contained in a twit in a particular time and clustering the keywords that describes same event by exploiting keywords' co-occurrence graph. Finally, we evaluates the validity of our method through experiments.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.29-30
• /
• 2019

본 논문에서는 다중 컨트롤러가 존재하는 분산 SDN 환경에서 과도한 제어 메시지로 인한 과부하된 컨트롤러의 부하를 줄이기 위하여 이주할 스위치를 K-means 군집화와 Harmony Search(HS)를 기반으로 선정 하는 기법을 제안하였다. 기존에 HS를 이용하여 이주할 스위치를 선택하는 기법이 제시되었으나, 시간 소모에 비하여 정확도가 부족한 단점이 있다. 또한 Harmony Memory(HM) 구축을 위해 메모리 소모 또한 크다. 이를 해결하기 위하여 본 논문에서는 유클리드 거리를 기반으로 하는 K-means 군집화를 이용하여 이주할 스위치를 골라내어 HM의 크기를 줄이고 이주 효율을 향상 시킨다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.5
• /
• pp.1223-1237
• /
• 2024

Autonomous vehicles use onboard sensors to sense the surrounding environment. In complex autonomous driving scenarios, the detection and recognition capabilities are constrained, which may result in serious accidents. An efficient way to enhance the detection and recognition capabilities is establishing collaborations with the neighbor vehicles. However, the collaborations introduce additional challenges in terms of the data heterogeneity, communication cost, and data privacy. In this paper, a novel personalized federated learning framework is proposed for addressing the challenges and enabling efficient collaborations in autonomous driving environment. For obtaining a global model, vehicles perform local training and transmit logits to a central unit instead of the entire model, and thus the communication cost is minimized, and the data privacy is protected. Then, the inference similarity is derived for capturing the characteristics of data heterogeneity. The vehicles are divided into clusters based on the inference similarity and a weighted aggregation is performed within a cluster. Finally, the vehicles download the corresponding aggregated global model and train a personalized model which is personalized for the cluster that has similar data distribution, so that accuracy is not affected by heterogeneous data. Experimental results demonstrate significant advantages of our proposed method in improving the efficiency of collaborative perception and reducing communication cost.

• Journal of KIISE:Software and Applications
• /
• v.33 no.8
• /
• pp.730-740
• /
• 2006

In recent years, bullied students and rogue groups in teenagers make many serious social problems. In this paper we propose a novel approach that more objectively analyzes peer relationships among students. As the data for objective analysis, we use communication history records that are collected from various communication tools such as telephones, e-mails, and messengers. We use the simple intuition that communication history records implicitly contain peer relationship information. And, we adopt data mining techniques for the more systematic analysis. The proposed peer relationship analysis consists of the following steps. First, we formally define the notion of degree of familiarity between friends, and present mathematical equations that compute the degree based on communication history records. In the proposed method, we use the intuition that the degree of familiarity from student x to student y becomes higher as x makes the more communications with y. Second, by using the degree of familiarity between students, we find out the students who are potentially bullied. This procedure is based on the assumption that a bullied student may have a very small number of history records from other students to him. Third, we adopt the clustering technique, one of the most representative data mining techniques, to find out meaningful student groups by using the degree of familiarity. To use the clustering technique, we formally define the notion of similarity between friends based on the degree of familiarity, and perform clustering using the notion. Last, to show the practicality of the proposed method, we have implemented the method and interpreted the meaning of the experimental results. Overall, we believe that our research result provides an effective framework that analyzes peer relationships more objectively and more systematically.

• Journal of KIISE:Software and Applications
• /
• v.31 no.4
• /
• pp.467-477
• /
• 2004

While recommender systems were used by a few E-commerce sites former days, they are now becoming serious business tools that are re-shaping the world of I-commerce. And collaborative filtering has been a very successful recommendation technique in both research and practice. But there are two problems in personalized recommender systems, it is First-Rating problem and Sparsity problem. In this paper, we solve these problems using the associative relation clustering and “Lift” of association rules. We produce “Lift” between items using user's rating data. And we apply Threshold by -cut to the association between items. To make an efficiency of associative relation cluster higher, we use not only the existing Hypergraph Clique Clustering algorithm but also the suggested Split Cluster method. If the cluster is completed, we calculate a similarity iten in each inner cluster. And the index is saved in the database for the fast access. We apply the creating index to predict the preference for new items. To estimate the Performance, the suggested method is compared with existing collaborative filtering techniques. As a result, the proposed method is efficient for improving the accuracy of prediction through solving problems of existing collaborative filtering techniques.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.11
• /
• pp.435-442
• /
• 2018

Outlier detection means to detect data samples that deviate significantly from the distribution of normal data. Most outlier detection methods calculate an outlier score that indicates the extent to which a data sample is out of normal state and determine it to be an outlier when its outlier score is above a given threshold. However, since the range of an outlier score is different for each data and the outliers exist at a smaller ratio than the normal data, it is very difficult to determine the threshold value for an outlier score. Further, in an actual situation, it is not easy to acquire data including a sufficient amount of outliers available for learning. In this paper, we propose a clustering-based outlier detection method by constructing a model representing a normal data region using only normal data and performing binary classification of outliers and normal data for new data samples. Then, by dividing the given normal data into chunks, and constructing a clustering model for each chunk, we expand it to the ensemble method combining the decision by the models and apply it to the streaming data with dynamic changes. Experimental results using real data and artificial data show high performance of the proposed method.