Go to the main menu
Skip to content
Go to bottom
REFERENCE LINKING PLATFORM OF KOREA S&T JOURNALS
> Journal Vol & Issue
The KIPS Transactions:PartB
Journal Basic Information
Journal DOI :
Korea Information Processing Society
Editor in Chief :
Volume & Issues
Volume 9B, Issue 6 - Dec 2002
Volume 9B, Issue 5 - Oct 2002
Volume 9B, Issue 4 - Aug 2002
Volume 9B, Issue 3 - Jun 2002
Volume 9B, Issue 2 - Apr 2002
Volume 9B, Issue 1 - Feb 2002
Selecting the target year
Real-time Hausdorff Matching Algorithm for Tracking of Moving Object
Chun, Quan ; Lee, Joo-Shin ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 707~714
DOI : 10.3745/KIPSTB.2002.9B.6.707
This paper presents a real-time Hausdorff matching algorithm for tracking of moving object acquired from an active camera. The proposed method uses the edge image of object as its model and uses Hausdorff distance as the cost function to identify hypothesis with the model. To enable real-time processing, a high speed approach to calculate Hausdorff distance and half cross matching method to improve performance of existing search methods are also presented. the experimental results demonstrate that the proposed method can accurately track moving object in real-time.
Design and Implementation of Real-time Moving Picture Encoder Based on the Fractal Algorithm
Kim, Jae-Chul ; Choi, In-Kyu ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 715~726
DOI : 10.3745/KIPSTB.2002.9B.6.715
In this paper, we construct real-time moving picture encoder based on fractal theory by using general purpose digital signal processors. The constructed encoder is implemented using two fixed-point general DSPs (ADSP2181) and performs image encoding by three stage pipeline structure. In the first pipeline stage, the image grabber acquires image data from NTSC standard image signals and stores digital image into frame memory. In the second stage, the main controller encode image dada using fractal algorithm. The last stage, output controller perform Huffman coding and result the coded data via RS422 port. The performance tests of the constructed encoder shows over 10 frames/sec encoding speed for QCIF data when all the frames are encoded. When we encode the images using the interframe and redundency based on the proposed algorithms, encoding speed increased over 30 frames/sec in average.
A Measurement Algorithm using Gray-level Thresholding in Automatic Refracto-Keratometer
Seong, Won ; Park, Jong-Won ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 727~734
DOI : 10.3745/KIPSTB.2002.9B.6.727
Currently. people become interested in the development of measuring instrument related to eyesight. In this study, we developed software of electronic part in automatic refracto-keratometer. If an automatic system, which uses images from an optical instrument, can inform the in-spector of an accurate eyesight measured value after the internal process, the frequency of mistakenly observed value will be reduced considerably. This software is using morphological filtering and gray-level signal enhancing techniques. The morphological filtering is the first process, from images of the optical instrument, to transform an original image which is hard to process into manageable one. The second process is a signal enhancing technique to the first processed image using gray -level thresholding technique and is used to reduce an error caused by the variety in distribution of the gray value of image. Therefore, this software system in electronic part will make more effective eyesight measurement by reducing the error effectively when applied to the optical image which is difficult to get accurate measurement value.
Improving Learning Performance of Support Vector Machine using the Kernel Relaxation and the Dynamic Momentum
Kim, Eun-Mi ; Lee, Bae-Ho ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 735~744
DOI : 10.3745/KIPSTB.2002.9B.6.735
This paper proposes learning performance improvement of support vector machine using the kernel relaxation and the dynamic momentum. The dynamic momentum is reflected to different momentum according to current state. While static momentum is equally influenced on the whole, the proposed dynamic momentum algorithm can control to the convergence rate and performance according to the change of the dynamic momentum by training. The proposed algorithm has been applied to the kernel relaxation as the new sequential learning method of support vector machine presented recently. The proposed algorithm has been applied to the SONAR data which is used to the standard classification problems for evaluating neural network. The simulation results of proposed algorithm have better the convergence rate and performance than those using kernel relaxation and static momentum, respectively.
A Study on the Guided Search Method for Transcoding MPEG2 P frame to H.263 P frame in a Compressed Domain
Um, Sung-Min ; Kang, Eui-Seon ; Lim, Young-Wan ; Hwang, Jae-Gak ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 745~752
DOI : 10.3745/KIPSTB.2002.9B.6.745
The purpose of the paper is to enable a format transcoding between a heterogeneous compression format in a real time, and to enhance the compression ratio using characteristics of the compressed frame. In this paper, for the heterogeneous format transcoding, we tried to transcode from MPEG2 having a lower compression ratio to H.263 having a higher compression ratio. After analyzing MPEG 2 bit stream and H.263 bit stream of the same original video, we found that the number of intra coded macro blocks in MPEG 2 data is much higher than the number of the intra coded macro blocks in H.263 data. In the process of P frame generation, a intra coded macro block is generated when a motion estimation value representing the similarity between the previous frame and current frame does not meet a threshold. Especially the intra coded macro block has a great impact on the compression ratio. Hence the paper, we tried to minimize the number of intra coded macro blocks in H.263 data stream which is transcoded from MPEG 2 in a compressed domain. For the purpose, we propose a guided search method for transcoding the INTRA coded block into INTER coded block using the information about motion vectors surrounding the intra macro block in order to minimize the complexity of the motion estimation process. The experimental results show that the transcoding of MPEG 2 into H.263 can be done in a real time successfully.
The Color Cross-Correlogram for the Robust Image Retrieval in the Size Change of Regions
An, Myoung-Seok ; Cho, Seok-Je ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 753~758
DOI : 10.3745/KIPSTB.2002.9B.6.753
This paper proposes the color Cross-correlogram and its extraction method for efficient image retrieval. Color cross-correlogram represents the probability that different colors are existed at any two pixels whose distance is fixed in an image. Color cross-correlogram doesn`t have the information about the region size that has a color, so color cross-correlogram can have good performance in retrieving images that have different size color regions. The experiments say that we can get the good retrieval results in the images that have various size color regions, and get the better retrieval results when using color cross-correlogram than those of retrieval using color correlogram.
A Machine Learning Approach to Web Image Classification
Cho, Soo-Sun ; Lee, Dong-Woo ; Han, Dong-Won ; Hwang, Chi-Jung ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 759~764
DOI : 10.3745/KIPSTB.2002.9B.6.759
Although image occupies a large part of importance on the Web documents, there have not been many researches for analyzing and understanding it. Many Web images are used for carrying important information but others are not used for it. In this paper classify the Web images from presently served Web sites to erasable or non-erasable classes. based on machine learning methods. For this research, we have detected 16 special and rich features for Web images and experimented by using the Baysian and decision tree methods. As the results, F-measures of 87.09%, 82.72% were achived for each method and particularly, from the experiments to compare the effects of feature groups, it has proved that the added features on this study are very useful for Web image classification.
Design and Implementation for Korean Character and Pen-gesture Recognition System using Stroke Information
Oh, Jun-Taek ; Kim, Wook-Hyun ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 765~774
DOI : 10.3745/KIPSTB.2002.9B.6.765
The purpose of this paper is a design and implementation for korean character and pen-gesture recognition system in multimedia terminal, PDA and etc, which demand both a fast process and a high recognition rate. To recognize writing-types which are written by various users, the korean character recognition system uses a database which is based on the characteristic information of korean and the stroke information Which composes a phoneme, etc. In addition. it has a fast speed by the phoneme segmentation which uses the successive process or the backtracking process. The pen-gesture recognition system is performed by a matching process between the classification features extracted from an input pen-gesture and the classification features of 15 pen-gestures types defined in the gesture model. The classification feature is using the insensitive stroke information. i.e., the positional relation between two strokes. the crossing number, the direction transition, the direction vector, the number of direction code. and the distance ratio between starting and ending point in each stroke. In the experiment, we acquired a high recognition rate and a fart speed.
Design of Adaptive Electronic Commerce Agents Using Machine Learning Techniques
Baek,, Hey-Jung ; Park, Young-Tack ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 775~782
DOI : 10.3745/KIPSTB.2002.9B.6.775
As electronic commerce systems have been widely used, the necessity of adaptive e-commerce agent systems has been increased. These kinds of agents can monitor customer`s purchasing behaviors, clutter them in similar categories, and induce customer`s preference from each category. In order to implement our adaptive e-commerce agent system, we focus on following 3 components-the monitor agent which can monitor customer`s browsing/purchasing data and abstract them, the conceptual cluster agent which cluster customer`s abstract data, and the customer profile agent which generate profile from cluster, In order to infer more accurate customer`s preference, we propose a 2 layered structure consisting of conceptual cluster and inductive profile generator. Many systems have been suffered from errors in deriving user profiles by using a single structure. However, our proposed 2 layered structure enables us to improve the qualify of user profile by clustering user purchasing behavior in advance. This approach enables us to build more user adaptive e-commerce system according to user purchasing behavior.
Fuzzy Pr/T Net Representation of Interval-valued Fuzzy Set Reasoning
Cho, Sang-Yeop ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 783~790
DOI : 10.3745/KIPSTB.2002.9B.6.783
This paper proposes a fuzzy Pr/T net representation of interval-valued fuzzy set reasoning, where fuzzy production rules are used for knowledge representation, and the belief of fuzzy production rules are represented by interval-valued fuzzy sets. The presented interval-valued fuzzy reasoning algorithm is much closer to human intuition and reasoning than other methods because this algorithm uses the proper belief evaluation functions according to fuzzy concepts in fuzzy production rules.
Korean Probabilistic Dependency Grammar Induction by morpheme
Choi, Seon-Hwa ; Park, Hyuk-Ro ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 791~798
DOI : 10.3745/KIPSTB.2002.9B.6.791
In this thesis. we present a new method for inducing a probabilistic dependency grammar (PDG) from text corpus. As words in Korean are composed of a set of more basic morphemes, there exist various dependency relations in a word. So, if the induction process does not take into account of these in-word dependency relations, the accuracy of the resulting grammar nay be poor. In comparison with previous PDG induction methods. the main difference of the proposed method lies in the fact that the method takes into account in-word dependency relations as well as inter-word dependency relations. To access the performance of the proposed method, we conducted an experiment using a manually-tagged corpus of 25,000 sentences which is complied by Korean Advanced Institute of Science and Technology (KAIST). The grammar induction produced 2,349 dependency rules. The parser with these dependency rules shoved 69.77% accuracy in terms of the number of correct dependency relations relative to the total number dependency relations for best-1 parse trees of sample sentences. The result shows that taking into account in-word dependency relations in the course of grammar induction results in a more accurate dependency grammar.
A Development of Hangul-Phone Keyboard and a Generation Method for Korean Nickname of Telephone Number
Choi, Jae-Hyuk ; Chung, Jae-Yeul ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 799~808
DOI : 10.3745/KIPSTB.2002.9B.6.799
With the enormous advance of communication technology, telephone numbers to be memorized is increased because most people have a lot of telephone numbers. Thus it needs a method for people to easily memorize those telephone numbers. A traditional method is to use Korean nickname by 4 syllable Korean words such as "Pal -li Pal-li," "Yi- sa Yi-sa" whose initial sounds are corresponding to Korean sounds of telephone numbers. However, this method does not apply to all the telephone numbers, and only apply to a few telephone numbers. This paper proposed a Hangul-phone keyboard (TELNICK) considering the reduction of finger`s moving distance and the number of strokes, and the generation of Korean nicknames for easy memorizing. This paper also proposes a method to generate many Korean nicknames that can correspond to 4-digit or 3-digit telephone numbers by using TELNICK keyboard and presents the form of Korean nicknames that can correspond to 7-digit or 8-digit telephone numbers.t telephone numbers.
Korean Probabilistic Syntactic Model using Head Co-occurrence
Lee, Kong-Joo ; Kim, Jae-Hoon ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 809~816
DOI : 10.3745/KIPSTB.2002.9B.6.809
Since a natural language has inherently structural ambiguities, one of the difficulties of parsing is resolving the structural ambiguities. Recently, a probabilistic approach to tackle this disambiguation problem has received considerable attention because it has some attractions such as automatic learning, wide-coverage, and robustness. In this paper, we focus on Korean probabilistic parsing model using head co-occurrence. We are apt to meet the data sparseness problem when we`re using head co-occurrence because it is lexical. Therefore, how to handle this problem is more important than others. To lighten the problem, we have used the restricted and simplified phrase-structure grammar and back-off model as smoothing. The proposed model has showed that the accuracy is about 84%.
Computational Processing of Korean Dialogue and the Construction of Its Representation Structure Based on Situational Information
Lee, Dong-Young ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 817~826
DOI : 10.3745/KIPSTB.2002.9B.6.817
In Korean dialogue honorification phenomenon may occur, an honorific pronoun may be used, and a subject or an object may be completely omitted when it can be recovered based on context. This paper proposes that in order to process Korean dialogue in which such distinct linguistic phenomena occur and to construct its representation structure we mark and use the following information explicitly, not implicitly : information about dialogue participants, information about the speech act of an utterance, information about the relative order of social status for the people involved in dialogue, and information flow among utterances of dialogue. In addition, this paper presents a method of marking and using such situational information and an appropriate representation structure of Korean dialogue. In this paper we set up Korean dialogue representation structure by modifying and extending DRT (Discourse Representation Theory) and SDRT (Segmented Discourse Representation Theory). Futhermore, this paper shows how to process Korean dialogue computationally and construct its representation structure by using Prolog programming language, and then applies such representation structure to spontaneous Korean dialogue to know its validity.
A Study on Creation of Hangeu-Romanization Conversion Table Using Petri-Nets
Kim, Kyung-Jing ; Choi, Young-Kyoo ; Rhee, Sang-Burm ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 827~834
DOI : 10.3745/KIPSTB.2002.9B.6.827
In this paper, we proposed the formation of Korean-Roman alphabet notation conversion table for the generation of Korean-Roman alphabet notation that also meets revised Roman alphabet notation. Introduced a mathematical analyzing method of the natural language which used a petrinet model so that a base of Roman alphabet notation analyzed standard pronunciation and Roman alphabet notation to work mathematically. It display the practical example through a petrinet modeling of a plan and Roman alphabet notation to create a Korean Roman alphabet notation conversion table with the method of the analysis that used a petrinet model, and present a mathematical modeling plan and application method of Korean. We developed application program based on window in order to verify a created Korean-Roman alphabet notation conversion table, and compared the result of an application program with Roman alphabet notation of an Roman alphabet notation example dictionary.
Design and Evaluation of a Channel Reservation Patching Method for True VOD Systems
Lee, Joo-Yung ; Ha, Sook-Jeong ; Bae, Ihn-Han ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 835~844
DOI : 10.3745/KIPSTB.2002.9B.6.835
The number of channels available to a video server is limited since the number of channels a video server can support is determined by its communication bandwidth. Several approaches such as batching, piggybacking and patching have been proposed to reduce I/O demand on the video server by sharing multicast data. Patching has been shown to be efficient in the matter of the cost for VOD systems. Unlike conventional multicast techniques, patching is a dynamic multicast scheme which enables a new request to join an ongoing multicast. In addition, true VOD can be achieved since a new request can be served immediately without having to wait for the next multicast. In this paper. we propose two types of channel reservation patching algorithm : a fixed channel reservation patching and a variable channel reservation patching. To immediately schedule the requests for popular videos, these algorithms reserve the channels of video server for the fixed number of popular videos or for the variable number of popular videos which is determined dynamically according to the load of video server. The performance of the proposed algorithms is evaluated through simulations, and compared with that of simple patching. Our performance measures are average defection rate, average latency, service fairness and the amount of buffered data according to video server loads. Simulation results show that the proposed channel reservation patching algorithms provide better performance compared to simple patching algorithm.
Design and Implementation of Middleware supporting translation of EDI using XML
Choi, Gwang-Mi ; Park, Su-Young ; Jung, Chai-Yeoung ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 845~852
DOI : 10.3745/KIPSTB.2002.9B.6.845
Electronic document processing using EDl (Electronic Data Interchange) must exchange documents using VAN (Value Added Network). However. the use of exclusive software needs alteration of a new document and the use of VAN(Value Added Network) demands an exchange of document and high cost for maintenance. Due to these problems, the existing EDI (Electronic Data Interchange) is turning into Web-based EDI (Electronic Data Interchange). This paper suggests techniques that change EDI (Electronic Data Interchange) messages which exist in two relational databases into XML (extensible Markeup Language) using the JDBC bridge. Also this paper proposes a method that recovers schema using converted XML (extensible Markeup Language) file, and a method which inserts an original record into a declared table. This solves the limitation of an original method that have to use sane database management system and also overcomes the problem in certain circumstances where the EDI (Electronic Data Interchange) exchange does not work.
Back-Propagation Neural Network Based Face Detection and Pose Estimation
Lee, Jae-Hoon ; Jun, In-Ja ; Lee, Jung-Hoon ; Rhee, Phill-Kyu ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 853~862
DOI : 10.3745/KIPSTB.2002.9B.6.853
Face Detection can be defined as follows : Given a digitalized arbitrary or image sequence, the goal of face detection is to determine whether or not there is any human face in the image, and if present, return its location, direction, size, and so on. This technique is based on many applications such face recognition facial expression, head gesture and so on, and is one of important qualify factors. But face in an given image is considerably difficult because facial expression, pose, facial size, light conditions and so on change the overall appearance of faces, thereby making it difficult to detect them rapidly and exactly. Therefore, this paper proposes fast and exact face detection which overcomes some restrictions by using neural network. The proposed system can be face detection irrelevant to facial expression, background and pose rapidily. For this. face detection is performed by neural network and detection response time is shortened by reducing search region and decreasing calculation time of neural network. Reduced search region is accomplished by using skin color segment and frame difference. And neural network calculation time is decreased by reducing input vector sire of neural network. Principle Component Analysis (PCA) can reduce the dimension of data. Also, pose estimates in extracted facial image and eye region is located. This result enables to us more informations about face. The experiment measured success rate and process time using the Squared Mahalanobis distance. Both of still images and sequence images was experimented and in case of skin color segment, the result shows different success rate whether or not camera setting. Pose estimation experiments was carried out under same conditions and existence or nonexistence glasses shows different result in eye region detection. The experiment results show satisfactory detection rate and process time for real time system.
A Method to Manage the Map for On-Line RPG Supporting Full 3D
Lee, Nam-Jae ; Kwak, Hoon-Sung ;
The KIPS Transactions:PartB, volume 9B, issue 6, 2002, Pages 863~868
DOI : 10.3745/KIPSTB.2002.9B.6.863
In this paper we suggested a map management method for on line RPG supporting full 3D. In general, it is possible to support full 3D with client engine. Rut the online game server cannot show the expected performance under normal price constrains with current hardware technology, since the amount of data for management of 3D on-line game server increase exponentially with respect to the size of game world. To solve this problem. we introduced the "special object" which makes it possible for on-line game server with low performance hardware. This suggested method gave concrete from to manage full 3D for server as well as clients.s clients.