• Journal of Information Management
• /
• v.28 no.3
• /
• pp.46-73
• /
• 1997

In these days, there are many studies to improve the effectiveness of retrieval systems. One method is to use thesaurus in the systems. But the establishment of the saurus independently is very difficult and requiers a great deal of time and labor. The goal of this research is to examine which problems there are. The texts used in the thesis were "Thesaurus of ERIC descriptors(7th ed.1977)" and "KEID Thesaurus(1981)" established by translation. The results of the thesis were as follows, First, there were many differences of equivalence classes of terms between English and Korean. Second, the errors of translation made words inadequate to use as a descriptors. Third, because of differences of traditions, thought and culture, associative relationship and hierarchical relation did not coincide between terns, Fourth, because the terms of the thesaurus of ERIC descriptors serve Western usages, the terms used in Korea were not or insufficient. So it way necessary to add our own therms. As mentioned above, translating a thesaurus for use as a documentation language for indexing and retrieval is not the optimum way of building up a thesaurus.

• Journal of Internet Computing and Services
• /
• v.8 no.5
• /
• pp.67-79
• /
• 2007

As the Web technologies and wired/wireless network are improved and various new multimedia services are introduced recently, need for searching multimedia including video data has been much increasing, The previous approaches for multimedia retrieval, however, do not make use of the relationships among semantic concepts contained in multimedia contents in an efficient way and provide only restricted search results, This paper proposes a multimedia retrieval system exploiting semantic relevancy of multimedia contents based on a domain ontology, We show the effectiveness of the proposed system by experiments on a prototype system we have developed. The proposed multimedia retrieval system can extend a given search keyword based on the relationships among the semantic concepts in the ontology and can find a wide range of multimedia contents having semantic relevancy to the input keyword. It also presents the results categorized by the semantic meaning and relevancy to the keyword derived from the ontology. Independency of domain ontology with respect to metadata on the multimedia contents is preserved in the proposed system architecture.

• Journal of the Korean Society for information Management
• /
• v.18 no.4
• /
• pp.183-200
• /
• 2001

The traditional way of access to image information is through descriptive keyword searching. However, many studies in image indexing and retrieval have reached a consensus on the difficulties and limitations of text-base image description. This research investigates the feasibility of using visual browsing that is being used comparatively much less than keyword searching. The effectiveness of Keyword access versus visual access were examined through experiments in which participants searched for pictures of specified emotions using different access modes: keywords only, visual browsing only, and the combination of both. It was found that keyword searching was appropriate for clean searches while visual browsing was the effective way to browse many pictures quickly, thus finding more relevant pictures. Findings and results can guide of images retrieval systems, especially the retrieval system of subjective and interpretive information.

• Journal of Broadcast Engineering
• /
• v.5 no.1
• /
• pp.2-9
• /
• 2000

This paper presents a statistical motion activity description method and video retrievals by using the intensity and directions of the extracted motion vectors from video sequence. Since the proposed method can represent temporal and spatial cognitive characteristics of an entire video, several images between key frames, and images in a certain interval, it can be effectively applied to digital video services such as video retrieval, surveilance, multimedia database, and broadcasting filterings. In the paper, the effectiveness of the proposed algorithm is shown with a lot of shots of MPEG-7 video dataset.

• The KIPS Transactions:PartD
• /
• v.18D no.5
• /
• pp.323-328
• /
• 2011

For content-based image retrieval, the earth mover's distance and the optimal color composition distance are proposed to measure the dissimilarity. Although providing good retrieval results, both methods are too time-consuming to be used in a large image database. To solve the problem, we propose a new distance function that calculates an approximate earth mover's distance in linear time. To calculate the dissimilarity in linear time, the proposed approach employs the space-filling curve. We have performed extensive experiments to show the effectiveness and efficiency of the proposed approach. The results reveal that our approach achieves almost the same results with the EMD in linear time.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.4300-4314
• /
• 2019

With the development of video-related applications, media content has increased dramatically through applications. There is a substantial amount of near-duplicate videos (NDVs) among Internet videos, thus NDVR is important for eliminating near-duplicates from web video searches. This paper proposes a novel NDVR system that supports large-scale retrieval and contributes to the efficient and accurate retrieval performance. For this, we extracted keyframes from each video at regular intervals and then extracted both commonly used features (LBP and HSV) and new image features from each keyframe. A recent study introduced a new image feature that can provide more robust information than existing features even if there are geometric changes to and complex editing of images. We convert a vector set that consists of the extracted features to binary code through a set of hash functions so that the similarity comparison can be more efficient as similar videos are more likely to map into the same buckets. Lastly, we calculate similarity to search for NDVs; we examine the effectiveness of the NDVR system and compare this against previous NDVR systems using the public video collections CC_WEB_VIDEO. The proposed NDVR system's performance is very promising compared to previous NDVR systems.

• Journal of the Korean Society for information Management
• /
• v.11 no.1
• /
• pp.31-56
• /
• 1994

To evaluate the retrieval effectiveness of the B03lean Request Conversion Mod4 the Relevance Weighted Boolean Model, and the P-NORM Model, the present study has been done with expenmental tests. It is proven that the Relevance Weighted Bdean Model is more effective in precision and the document output ranks than the other ones. The expenmental results indmte a promisii application of relevance mformation and weigh- schemes.

• Journal of Korea Society of Industrial Information Systems
• /
• v.9 no.4
• /
• pp.26-31
• /
• 2004

Due to the recent explosive progress of Web, We can easily access a large number of images from m. In this paper, we describe our approach of developing a general purpose content based image retrieval system over the H using a Web agent. The Web agent extracts text information of images from the links and file contents in HTML. The proposed system retrieves the images from database using the query by sketch and the query by example on Web browser. Experimental results demonstrate the effectiveness of the new approach.

• Journal of Korea Multimedia Society
• /
• v.14 no.12
• /
• pp.1544-1548
• /
• 2011

A process of choosing a subset of original features, so called feature selection, is considered as a crucial preprocessing step to image processing applications. There are already large pools of techniques developed for machine learning and data mining fields. In this paper, basically two methods, non-feature selection and feature selection, are investigated to compare their predictive effectiveness of classification. Color co-occurrence feature is used for defining image features. Standard Sequential Forward Selection algorithm are used for feature selection to identify relevant features and redundancy among relevant features. Four color spaces, RGB, YCbCr, HSV, and Gaussian space are considered for computing color co-occurrence features. Gray-level image feature is also considered for the performance comparison reasons. The experimental results are presented.

• Journal of Korean Library and Information Science Society
• /
• v.22
• /
• pp.1-26
• /
• 1995

Indexes, especially subject indexes, are major tools for information retrieval. To enhance the retrieval effectiveness of subject indexes, the semantic and syntactic relationships of indexing languages are very important elements. This paper examines the afore-mentioned relationships, based on purely the syntax and semantics of Korean language. The outlines of this study are as follows: 1. The characteristics and usages of controlled vocabularies, particularly subject headings lists and thesaury, are reviewed. 2. The semantic relationships, such as equivalence, hierarchical and associative relationships, are defined, and their categories are investigated in detail. Accordingly, the usages of 'See' and 'See also' references are suggested circumstantially. 3. The syntactic relationships are also examined. Particularly, for the syntactic relationships of multiword indexing terms, two kinds of subject entry formats are compared. Since it is more rational for subject headings organized by the principle of context-dependency, the two-fine entry format is recommended for subject indexes. 4. Computerized production techniques of 'See' and 'See also' reference for the semantic relationships of indexing terms are presented. 5. Computerized production techniques of subject indexes representing the syntactic relationships of indexing terms are also presented.