• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.1
• /
• pp.249-268
• /
• 2014

Relevance feedback is an effective tool to bridge the gap between superficial image contents and medically-relevant sense in content-based medical image retrieval. In this paper, we propose an interactive medical image search framework based on pairwise constraint propagation. The basic idea is to obtain pairwise constraints from user feedback and propagate them to the entire image set to reconstruct the similarity matrix, and then rank medical images on this new manifold. In contrast to most of the algorithms that only concern manifold structure, the proposed method integrates pairwise constraint information in a feedback procedure and resolves the small sample size and the asymmetrical training typically in relevance feedback. We also introduce a long-term feedback strategy for our retrieval tasks. Experiments on two medical image datasets indicate the proposed approach can significantly improve the performance of medical image retrieval. The experiments also indicate that the proposed approach outperforms previous relevance feedback models.

• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.612-615
• /
• 2006

The management of image information settles as an important field with the advent of multimedia age and we are in need of the effective retrieval method to manage systematically image information. It is used to color, texture, and shape features for content-based image retrieval. And existing methods using multiple features have problems that the retrieval process is embarrassed because each weight is set up manually. So we have solved these problems by assignment of weight applying fuzzy integral. This paper proposed the optimized values of fuzzy measure by experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.3
• /
• pp.670-684
• /
• 2024

3D Cross-Modal Retrieval (3DCMR) is a task that retrieves 3D objects regardless of modalities, such as images, meshes, and point clouds. One of the most prominent methods used for 3DCMR is the Cross-Modal Center Loss Function (CLF) which applies the conventional center loss strategy for 3D cross-modal search and retrieval. Since CLF is based on center loss, the center features in CLF are also susceptible to subtle changes in hyperparameters and external inferences. For instance, performance degradation is observed when the batch size is too small. Furthermore, the Mean Squared Error (MSE) used in CLF is unable to adapt to changes in batch size and is vulnerable to data variations that occur during actual inference due to the use of simple Euclidean distance between multi-modal features. To address the problems that arise from small batch training, we propose a Noisy Center Loss (NCL) method to estimate the optimal center features. In addition, we apply the simple Siamese representation learning method (SimSiam) during optimal center feature estimation to compare projected features, making the proposed method robust to changes in batch size and variations in data. As a result, the proposed approach demonstrates improved performance in ModelNet40 dataset compared to the conventional methods.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.9 no.1
• /
• pp.114-121
• /
• 2008

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. In this paper, we propose semantic-based video retrieval system for Environment Education Contents 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 form query, the 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. From experiment, the designed and implemented system showed high precision ratio in performance assessment more than 90 percents.

• Journal of The Korean Association of Information Education
• /
• v.12 no.3
• /
• pp.313-322
• /
• 2008

As cyber education using e-Learning system is expanded, various Learning Contents has been created to satisfy the demand of various students. But since it takes long time and spends high cost to create e-Learning contents, it is hard to satisfy the demand of various students. To solve this problem standardizations of the Learning Contents and researches of Learning Contents Metadata are focused in reusability of Learning Contents and information retrieval of Learning Contents. In this paper to improve manageability and retrievability, legacy version control and document management system are introduced. Based on existing version control and document management system, we developed automated history control system. To conveniently provide retrieval, inquiry and integrating of Learning Contents, we researched Learning Contents Management System based on SOA to easily approach with the Learning Contents Management Server which is dispersed on wide area

• The Journal of the Korea Contents Association
• /
• v.10 no.8
• /
• pp.130-137
• /
• 2010

An advancement and wide use of portable multimedia devices make it easy to spread digital sound source, enabling users to use portable multimedia devices and search and enjoy digital sound sources and related contents. In opposition the off-line musical service market with CD stagnates gradually and in the crisis where the off-line musical service market vanishes away. In this paper, we propose the mobile music album information retrieval system which combines the barcode and a cellular phone musical service. To get digital sound sources and related information from music albums, the proposed retrieval system uses one-dimensional barcodes attached to music albums and obtains information on related albums from the sound source servers. Also, to reflect user preferences in search results, the system uses search frequency by music album, and searches albums in music genres selected by users in the order of the recent preference. The proposed mobile music album retrieval system is implemented on the basis of WIPI, and is now providing pilot services.

• The Journal of the Korea Contents Association
• /
• v.10 no.2
• /
• pp.490-498
• /
• 2010

In the web or not, all academic information have the creator which produces that information. The creator can be individual, organization, institution, or country. Most information consist of the title, author and content. The article among academic information is described by title, author, keywords, abstract, publisher, ISSN(International Standard Serial Number) and etc., and the patent information is consisted some metadata such as invention title, applicant, inventors, agents, application number, claim items etc. Most web-based academic information services provide search functions to user by processing and handling these metadata, and the search function using the author field is important. In this paper, we propose an effective indexing management for person name search, and search techniques using boosting factor and near operation based on phrase search to improve precision rate of search result. And we describe person name retrieval result with another expression name, co-authors and persons in same research field. The approach presented in this paper provides accurate data and additional search results to user efficiently.

• Journal of the Korean Society for information Management
• /
• v.33 no.2
• /
• pp.77-102
• /
• 2016

The effective searching for performing arts contents can be achieved by providing various access points and searching methods based on specialized metadata. The purpose of this study is to develop a faceted navigation interface which user could effectively and efficiently retrieve performing arts contents even if the users do not know accurate descriptive information about them. Therefore, the study, first, investigated search access points and navigation items providing by other the performing arts retrieval systems and to analyze information seeking behaviors of university students who major in music, dance and theater. And then, the study proposed the 36 facets with the 9 main facet categories suitable for performing arts and also proposed 27 descriptive elements suitable for performing arts contents. Finally, the study developed the performing arts contents retrieval system based faceted navigation interface with 3,360 experimental data and conducted an in-depth interview in terms of usability, serendipity, and efficiency. The applying the faceted navigations for searching performing arts contents will help users access and utilize them in the retrieval system and moreover satisfy user demands.

• Journal of Internet Computing and Services
• /
• v.2 no.1
• /
• pp.49-61
• /
• 2001

Internet has been becoming widely popular and making rapid progress and network technologies is showing extension in data transmission speeds. Rapid and convenient multimedia services supplied with high quality and high speed are being needed, This paper treats of the design and implement method of clip-based video retrieval system on the world-wide-web environments. The implemented system consists of the content-based indexing system supporting convenient services for video contents providers and the web-based retrieval system in order to make it easy and various information retrieval for users on the world-wide-web. Three important methods were used in the content-based indexing system. Key frame extracting method by dividing video data, clip file creation method by clustering related information and video database build method by using clip unit, In web-based retrieval system, retrieval method by using a key word, two dimension browsing method of key frame and real-time display method of the clip were used. As a result. the proposed methodologies showed a usefulness of video content providing. and provided an easy method for searching intented video content.

• Journal of Intelligence and Information Systems
• /
• v.15 no.2
• /
• pp.49-59
• /
• 2009

Contents-based image retrieval methods are in general more objective and effective than text-based image retrieval algorithms since they use color and texture in search and avoid annotating all images for search. SIM(Self-organizing Image browsing Map) is one of contents-based image retrieval algorithms that uses only browsable mapping results obtained by SOM(Self Organizing Map). However, SOM may have an error in selecting the right BMU in learning phase if there are similar nodes with distorted color information due to the intensity of light or objects' movements in the image. Such images may be mapped into other grouping nodes thus the search rate could be decreased by this effect. In this paper, we propose an improved SIM that uses HSV color model in extracting image features with color quantization. In order to avoid unexpected learning error mentioned above, our SOM consists of two layers. In learning phase, SOM layer 1 has the color feature vectors as input. After learning SOM Layer 1, the connection weights of this layer become the input of SOM Layer 2 and re-learning occurs. With this multi-layered SOM learning, we can avoid mapping errors among similar nodes of different color information. In search, we put the query image vector into SOM layer 2 and select nodes of SOM layer 1 that connects with chosen BMU of SOM layer 2. In experiment, we verified that the proposed SIM was better than the original SIM and avoid mapping error effectively.