Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
Journal of Korea Multimedia Society
Journal Basic Information
Journal DOI :
Korea Multimedia Society
Editor in Chief :
Volume & Issues
Volume 9, Issue 12 - Dec 2006
Volume 9, Issue 11 - Nov 2006
Volume 9, Issue 10 - Oct 2006
Volume 9, Issue 9 - Sep 2006
Volume 9, Issue 8 - Aug 2006
Volume 9, Issue 7 - Jul 2006
Volume 9, Issue 6 - Jun 2006
Volume 9, Issue 5 - May 2006
Volume 9, Issue 4 - Apr 2006
Volume 9, Issue 3 - Mar 2006
Volume 9, Issue 2 - Feb 2006
Volume 9, Issue 1 - Jan 2006
Selecting the target year
An Incremental Web Document Clustering Based on the Transitive Closure Tree
Youn Sung-Dae ; Ko Suc-Bum ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 1~10
In document clustering methods, the k-means algorithm and the Hierarchical Alglomerative Clustering(HAC) are often used. The k-means algorithm has the advantage of a processing time and HAC has also the advantage of a precision of classification. But both methods have mutual drawbacks, a slow processing time and a low quality of classification for the k-means algorithm and the HAC, respectively. Also both methods have the serious problem which is to compute a document similarity whenever new document is inserted into a cluster. A main property of web resource is to accumulate an information by adding new documents frequently. Therefore, we propose a new method of transitive closure tree based on the HAC method which can improve a processing time for a document clustering, and also propose a superior incremental clustering method for an insertion of a new document and a deletion of a document contained in a cluster. The proposed method is compared with those existing algorithms on the basis of a pre챠sion, a recall, a F-Measure, and a processing time and we present the experimental results.
Face Detection Based on Distribution Map
Cho Han-Soo ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 11~22
Recently face detection has actively been researched due to its wide range of applications, such as personal identification and security systems. In this paper, a new face detection method based on the distribution map is proposed. Face-like regions are first extracted by applying the skin color map with the frequency to a color image and then, possible eye regions are determined by using the pupil color distribution map within the face-like regions. This enables the reduction of space for finding facial features. Eye candidates are detected by means of a template matching method using weighted window, which utilizes the correlation values of the luminance component and chrominance components as feature vectors. Finally, a cost function for mouth detection and location information between the facial features are applied to each pair of the eye candidates for face detection. Experimental results show that the proposed method can achieve a high performance.
Performance Improvement of Continuous Digits Speech Recognition Using the Transformed Successive State Splitting and Demi-syllable Pair
Seo Eun-Kyoung ; Choi Gab-Keun ; Kim Soon-Hyob ; Lee Soo-Jeong ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 23~32
This paper describes the optimization of a language model and an acoustic model to improve speech recognition using Korean unit digits. Since the model is composed of a finite state network (FSN) with a disyllable, recognition errors of the language model were reduced by analyzing the grammatical features of Korean unit digits. Acoustic models utilize a demisyllable pair to decrease recognition errors caused by inaccurate division of a phone or monosyllable due to short pronunciation time and articulation. We have used the K-means clustering algorithm with the transformed successive state splitting in the feature level for the efficient modelling of feature of the recognition unit. As a result of experiments, 10.5% recognition rate is raised in the case of the proposed language model. The demi-syllable fair with an acoustic model increased 12.5% recognition rate and 1.5% recognition rate is improved in transformed successive state splitting.
Face Recognition Using First Moment of Image and Eigenvectors
Cho Yong-Hyun ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 33~40
This paper presents an efficient face recognition method using both first moment of image and eigenvector. First moment is a method for finding centroid of image, which is applied to exclude the needless backgrounds in the face recognitions by shitting to the centroid of face image. Eigenvector which are the basis images as face features, is extracted by principal component analysis(PCA). This is to improve the recognition performance by excluding the redundancy considering to second-order statistics of face image. The proposed methods has been applied to the problem for recognizing the 60 face images(15 persons *4 scenes) of 320*243 pixels. The 3 distances such as city-block, Euclidean, negative angle are used as measures when match the probe images to the nearest gallery images. In case of the 45 face images, the experimental results show that the recognition rate of the proposed methods is about 1.6 times and its the classification is about 5.6 times higher than conventional PCA without preprocessing. The city-block has been relatively achieved more an accurate classification than Euclidean or negative angle.
An Adaptive Z-buffer Algorithm for PDA Platform
Kim Dae-Young ; Kim Hyo-Chul ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 41~50
This paper aims to improve the efficiency of a 3D-graphic software engine in a PDA platform and the performance of a rasterizer. There are many problems in implementing a software engine in a mobile platform, due to its relatively weak processing power. Taking the advantages and complementing weaknesses of the depth-sort algorithm and the Z-buffer algorithm, we implemented an adaptive Z-buffer algorithm and proved its performance on several PDA platforms. Our algorithm was evaluated to have midway speed compared with two conventional algorithms. It also removed reversal errors in comparison with the depth-sort algorithm.
Performance of MC-CDMA Based UWB System
Kim Cheol-Soon ; Kwak Kyung-Sup ; Lee Hyung-Ki ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 51~58
In this paper, we evaluated Ultra Wideband MC-CDMA system which is the combination of DS-CDMA and OFDM systems, which have been drafting in current standardization for IEEE802.15.3a. Too many Rake Fingers are demanded in the DS-CDMA system to detect multi-path signals, which results in high system complexity. OFDM system fails to qualify for FCC certification unless frequency hopping is off. MC-CDMA has lower complexity compared to DS-CDMA and shows good performance against frequency selective fading. In addition, for a wide-band communication, less radio power per spectrum is allowed in the MC-CDMA system than in an OFDM system. The MC-CDMA system is analyzed by a numerical formula and compared with DS-CDMA and OFDM by a computer simulation.
A Dynamic Frame-skipping Scheme for Transcoder Using the Prediction Period
Kim Sung-Min ; Kim Young-Ju ; Park Seong-Ho ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 59~72
The rapid growth of technology for network and terminal devices is able to service multimedia contents in mobile environments. But as the environment of networks and the performance of devices is varied, the same quality service is very difficult to be provided for user in internet and mobile environment. Therefore, multimedia streaming system must provide an adaptive service in considering network and device environment. Video transcoding techniques are the good solution that can provide video streaming service in various network environments adaptively. This paper proposes a dynamic frame-skipping transcoding scheme to provide adaptive streaming service in various network environments. This scheme uses the prediction period for transcoding. A proposed transcoding scheme can reduce time complexity through combination of prediction period and skipping period to control frame-skipping rate in encoder. In simulation results, the performance of proposed scheme is similar to the performance of a traditional dynamic frame-skipping scheme in PSNR value. But the performance of proposed scheme outperforms the performance of traditional scheme in time complexity.
An Extended SIP-based Network Architecture and Session Management Methods for Internet QoS
Shin Kang-Sik ; Lee Jong-Hoon ; Hong Kyung-Pyo ; Yoo Sang-Jo ; Cho Eun-Hee ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 73~88
With the development of Next Generation convergence Network, the current Internet needs a new network architecture and new session establishment methods, which can support various applications requiring differentiated traffic processing and high quality of service instead of the conventional best-effort service. It is expected that, in next generation network based on All-IP. SIP will play an important role in the session establishment between the end to end user terminals. In this paper, we propose a new network architecture that can provide quality of service based on SIP using the QoS enabled SIP servers and policy servers, and we also propose a new session establishment method that apply the user's demands participating the session establishment when setting up the session using the SIP. Through simulations under various environments, we show the parameter ranges in terms of server capacity and the number of users in access networks to achieve reasonable delay bounds.
Connection Control and Network Management of OBS with GSMP Open Interface
Choi In-Sang ; Kim Choon-Hee ; Cha Young-Wook ; Kwon Tae-Hyun ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 89~100
The introduction of GSMP open interface to OBS network can materialize to separate the transport plane and the control plane in OBS network. This makes the implementation of OBS switches simple and provides various flexibility. However, the introduction of open interface will cause the connection setup delay because of the additional processing overhead of open interface protocol. Also, in GSMP based network, the location of network management functions are not defined explicitly and the research result about the OBS network management is almost nothing. This paper proposes a parallel connection setup mechanism using centralized connection control server to minimize connection setup delay in OBS network with GSMP open interface and defines managed objects to support connection, configuration, performance, and fault management for the management of OBS network with GSMP open interface. This paper also proposes a distributed network management model, in which the above managed objects are distributed in a controller and an OBS switch according to network management functions. We verify the possibility of OBS control and network management by implementing network management function using proposed parallel connection setup mechanism and distributed network management model.
The APIs Design for the Database Management of the Network Game Using Design Patterns
Kim Jong-Soo ; Kwon Oh-Jun ; Kim Tai-Suk ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 101~108
Developing a network game that is serviced on the internet requires a lot of work, time and manpower. Therefore, it is important to design a game in such a way so that existing design codes could be used. The database for the multi-player network game is accessed from many clients. To manage the resource effectively, it is required to design the APIs to be minimized the database access and to be dealt with the related data efficiently. For this, it is needed to apply the object-oriented design for the related software modules. In the paper, we propose the design patterns of GoF to implement the APIs thatis needed to deal with the database resource. The database design through the analysis of some game's GUI has the advantage to create the optimized schema set more quickly, because it minimize the review step and the modification step of the database design generally. In addition, we apply to the design patterns of GoF for the APIs design of the server-side database. These reduce the times of the program call between the game server and the database server. These also make easily the maintenance for the already developed APIs, and it makes easily the addition of new APIs.
Environment Map Based Disparity
Ryoo Seung-Taek ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 109~118
In this paper, we suggest the environment based disparity method that calculate the depth value of the objects from environment maps. This method using the disparity of the environment map can calculate the depth value from two environment map that acquire at different viewpoint. This method can decide the visibility of the object whether it is occluded others or not. Also, we can analogize the depth value of the object that does not relate the reference plane(in case of being in the air) and make three dimensional environment model using the proposed method
Task Scheduling Algorithm in Multiprocessor System Using Genetic Algorithm
Kim Hyun-Chul ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 119~126
The task scheduling in multiprocessor system is one of the key elements in the effective utilization of multiprocessor systems. The optimal assignment of tasks to multiprocessor is, in almost practical cases, an NP-hard problem. Consequently algorithms based on various modern heuristics have been proposed for practical reason. This paper proposes a new task scheduling algorithm using Genetic Algorithm which combines simulated annealing (GA+SA) in multiprocessor environment. In solution algorithms, the Genetic Algorithm (GA) and the simulated annealing (SA) are cooperatively used. In this method, the convergence of GA is improved by introducing the probability of SA as the criterion for acceptance of new trial solution. The objective of proposed scheduling algorithm is to minimize makespan. The effectiveness of the proposed algorithm is shown through simulation studies. In simulation studies, the result of proposed algorithm is better than that of any other algorithms.
A Semantic-based Video Retrieval System Using the Automatic Indexing Agent
Kim Sam-Keun ; Lee Jong-Hee ; Yoon Sun-Hee ; Lee Keun-Soo ; Seo Jeong-Min ;
Journal of Korea Multimedia Society, volume 9, issue 1, 2006, Pages 127~137
In order to process video data effectively, it is required that the content information of video data is loaded in database and semantic- based retrieval method can be available for various query of users. Currently existent contents-based video retrieval systems search by single method such as annotation-based or feature-based retrieval, and show low search efficiency and requires many efforts of system administrator or annotator form less perfect automatic processing. In this paper, we propose semantic-based video retrieval system which support semantic retrieval of various users by feature-based retrieval and annotation-based retrieval of massive video data. By user's fundamental query and selection of image for key frame that extracted from query, the automatic indexing agent gives the detail shape for annotation of extracted key frame. Also, key frame selected by user become query image and searches the most similar key frame through feature based retrieval method that propose. Therefore, we propose the system that can heighten retrieval efficiency of video data through semantic-based retrieval.