• Journal of Convergence for Information Technology
• /
• v.11 no.10
• /
• pp.45-52
• /
• 2021

In this paper, we proposed an Atrous Residual U-Net (AR-UNet) to improve the segmentation accuracy of semantic segmentation method based on U-Net. The U-Net is mainly used in fields such as medical image analysis, autonomous vehicles, and remote sensing images. The conventional U-Net lacks extracted features due to the small number of convolution layers in the encoder part. The extracted features are essential for classifying object categories, and if they are insufficient, it causes a problem of lowering the segmentation accuracy. Therefore, to improve this problem, we proposed the AR-UNet using residual learning and ASPP in the encoder. Residual learning improves feature extraction ability and is effective in preventing feature loss and vanishing gradient problems caused by continuous convolutions. In addition, ASPP enables additional feature extraction without reducing the resolution of the feature map. Experiments verified the effectiveness of the AR-UNet with Cityscapes dataset. The experimental results showed that the AR-UNet showed improved segmentation results compared to the conventional U-Net. In this way, AR-UNet can contribute to the advancement of many applications where accuracy is important.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.11
• /
• pp.1133-1144
• /
• 2019

Recently, the amount of image on the Internet has rapidly increased, due to the advancement of smart devices and various approaches to effective image retrieval have been researched under these situation. Existing image retrieval methods simply detect the objects in a image and carry out image retrieval based on the label of each object. Therefore, the semantic gap occurs between the image desired by a user and the image obtained from the retrieval result. To reduce the semantic gap in image retrievals, we connect the module for multiple objects classification based on deep learning with the module for human behavior classification. And we combine the connected modules with a behavior ontology. That is to say, we propose an image retrieval system considering the relationship between objects by using the combination of deep learning and behavior ontology. We analyzed the experiment results using walking and running data to take into account dynamic behaviors in images. The proposed method can be extended to the study of automatic annotation generation of images that can improve the accuracy of image retrieval results.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.10 no.1
• /
• pp.93-100
• /
• 2016

This study analyzed research trends of assistive devices used to prevent and support disability and increase convenience for people with disabilities in their daily lives. As a result, related research has continuously increased, and the research area has expanded from survey research to service-related research and assistive device development research. In addition, with regard to the use of related terms, terms such as 'rehabilitation engineering' and 'rehabilitation assistive technology' were used in the early stages. Currently, terms such as 'assistive technology', 'assistive technology service', 'assistive technology device', and 'assistive device' have been mainly used.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.991-1005
• /
• 2022

Satellite imaging is an effective supplementary data source for detecting and verifying nuclear activity. It is also highly beneficial in regions with limited access and information, such as nuclear installations. Time series analysis, in particular, can identify the process of preparing for the conduction of a nuclear experiment, such as relocating equipment or changing facilities. Differences in the semantic segmentation findings of time series photos were employed in this work to detect changes in meaningful items connected to nuclear activity. Building, road, and small object datasets made of KOMPSAT 3/3A photos given by AIHub were used to train deep learning models such as U-Net, PSPNet, and Attention U-Net. To pick relevant models for targets, many model parameters were adjusted. The final change detection was carried out by including object information into the first change detection, which was obtained as the difference in semantic segmentation findings. The experiment findings demonstrated that the suggested approach could effectively identify altered pixels. Although the suggested approach is dependent on the accuracy of semantic segmentation findings, it is envisaged that as the dataset for the region of interest grows in the future, so will the relevant scope of the proposed method.

• Korean Association of Arts Management
• /
• no.58
• /
• pp.205-239
• /
• 2021

This study explores the importance of semantic web-based network construction in art data archiving, as well as its meaning and value in the context of arts management along with its potential for future application. The study focuses on oral history obtained from the Arko Arts Archives that contained records of the lives and artistic views of early artists. In this study, the possibility of applying semantic web-based technology to materials concerning culture and the arts was discussed in five aspects based on the results of the case analysis. First, checking the relationship and discovering hidden artists are possible by revealing relationships between characters. Second, understanding and studying society and culture at a given time is possible by interpreting the contextual meaning of information. Third, art exploration can be done broadly and deeply, encompassing various genres from the perspective of the consumer. Fourth, through art construction, history can be reconstructed using a new and rich method. Fifth, expanding the scope beyond the boundaries of art is possible through convergence and collaboration of programs that handle big data. The network data can be used in various methods, such as art history research, art planning, and creation, throughout the art ecosystem. The results of the study suggest that digitizing a large quantity of data concerning culture and the arts is meaningful in arts management as well as identifying and analyzing the relationship network among data clusters using semantic web-based technology.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.10
• /
• pp.23-33
• /
• 2014

The development of the information retrieval field is evolving to the research field searching accurately for the information from thing finding rapidly a large amount of information. Personalization and the semantic web technology is a key technology. The automatic indexing technology about the web document and throughput go beyond the research stage and show up as the practical service. However, there is a lack of research on the document information retrieval field about the attached document type of except the web document. In this paper, we illustrate about the method in which it analyzed the text content of the unstructured documents prepared in the text, word, hwp form and it how to construction OWL ontology. To build TBox of the document ontology and the resources which can be obtained from the document is selected, and we implement with the system in order to utilize as the instant of the constructed document ontology. It is effectually usable in the information retrieval and document management system using the semantic technology of the correspondence document as the ontology automatic construction of this kind of the unstructured documents.

• Journal of Internet Computing and Services
• /
• v.17 no.4
• /
• pp.61-68
• /
• 2016

In this paper, an implementation method of augmented reality (AR) technology using Web 3.0 information services in the field of disaster response is proposed. The structure and characteristics of semantic web-based Web 3.0 are realized and a AR based mobile visual search (MVS) applied in the disaster sites is described. Based on Web 3.0 and AR MVS, a semantic web ontology oriented configuration scheme for disaster-related information and the communication scheme of information provided by AR technology are proposed. For the purpose of providing disaster-related and customized information to the disaster response site quickly and accurately, a method of leveraging Web 3.0 information services in AR technology is presented.

• Journal of the HCI Society of Korea
• /
• v.1 no.1
• /
• pp.9-20
• /
• 2006

This paper describes a framework for multimodal object manipulation in virtual environments. The gist of the proposed framework is the semantic integration of multimodal input using spatial ontology and user context to integrate the interpretation results from the inputs into a single one. The spatial ontology, describing the spatial relationships between objects, is used together with the current user context to solve ambiguities coming from the user's commands. These commands are used to reposition the objects in the virtual environments. We discuss how the spatial ontology is defined and used to assist the user to perform object placements in the virtual environment as it will be in the real world.

• Annual Conference on Human and Language Technology
• /
• 1997.10a
• /
• pp.367-374
• /
• 1997

This paper attempts to analyze Korean proverbs with pap 'rice' which plays an important role in the Korean community. I examine to analyze the data under the various frameworks: Grice, lakoff and Langacker. Proverbs use the contrast in order to focus the speaker's intention and to get the convince from hearers. I limited to analyze coordinate sentences since these distinctively show the contrast and the relation. In terms of the contrast and the relation, the semantic interpretations of pap can be easily taken. These semantic interpretations are classified under the Lakoff's metaphors.

• Journal of Computing Science and Engineering
• /
• v.5 no.4
• /
• pp.346-353
• /
• 2011

With the advent of ubiquitous computing, a massive amount of trajectory data has been published and shared in many websites. This type of computing also provides motivation for online mining of trajectory data, to fit user-specific preferences or context (e.g., time of the day). While many trajectory clustering algorithms have been proposed, they have typically focused on offline mining and do not consider the restrictions of the underlying road network and selection conditions representing user contexts. In clear contrast, we study an efficient clustering algorithm for Boolean + Clustering queries using a pre-materialized and summarized data structure. Our experimental results demonstrate the efficiency and effectiveness of our proposed method using real-life trajectory data.