• Journal of Digital Convergence
• /
• v.11 no.1
• /
• pp.165-172
• /
• 2013

Website creations and their renewal process are very difficult due to ever-changing mobile environments such as various operating systems and mobile devices. In this regard, HTML5 serves as the solution and guideline to developing well-functioning websites. Mobile websites using HTML5 can provide consistent contents and services even though users access them through various OS or device types. This paper examines a possibility of the HTML5-based website for a local private college located in Busan. For this, location-based map service and semantic auto dialing webform service have been implemented. After making the website available, students and staff members were satisfied with the improvement in the speed of loading time and error free contents service. In the future, the rest of HTML5 functionalities are planned to be implemented sequentially.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.1170-1178
• /
• 2022

Indoor POI data, an essential component of indoor spatial data, has attribute information of a specific place in the room and is the most critical information necessary for the user. Currently, indoor POI data is manually updated by direct investigation, which is expensive and time-consuming. Recently, research on updating POI using the attribute information of indoor photographs has been advanced to overcome these problems. However, the range of use, such as using only photographs with text information, is limited. Therefore, in this study, and to improvement this, I proposed a new method to update indoor POI data using a smartphone camera. In the proposed method, the POI name is obtained by classifying the indoor scene's photograph into artificial intelligence technology CNN and matching the location criteria to indoor spatial data through AR technology. As a result of creating and experimenting with a prototype application to evaluate the proposed method, it was possible to update POI data that reflects the real world with high accuracy. Therefore, the results of this study can be used as a complement or substitute for the existing methodologies that have been advanced mainly by direct research.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.8
• /
• pp.43-50
• /
• 2009

Text information in a digital video provides crucial information to acquire semantic information of the video. In the proposed method. text candidate regions are extracted from input sequence by using characteristics of stroke and text candidate regions are localized by using projection to produce text bounding boxes. Bounding boxes containing text regions are verified geometrically and each bounding box existing same location is tracked by calculating matching measure. which is defined as the mean of absolute difference between bounding boxes in the current frame and previous frames. Finally. text regions are enhanced using temporal redundancy of bounding boxes to produce final results. Experimental results for various videos show the validity of the proposed method.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_3
• /
• pp.939-951
• /
• 2022

The most important factor in designing autonomous driving systems is to recognize the exact location of the vehicle within the surrounding environment. To date, various sensors and navigation systems have been used for autonomous driving systems; however, all have limitations. Therefore, the need for high-definition (HD) maps that provide high-precision infrastructure information for safe and convenient autonomous driving is increasing. HD maps are drawn using three-dimensional point cloud data acquired through a mobile mapping system (MMS). However, this process requires manual work due to the large numbers of points and drawing layers, increasing the cost and effort associated with HD mapping. The objective of this study was to improve the efficiency of HD mapping by segmenting semantic information in an MMS point cloud into six classes: roads, curbs, sidewalks, medians, lanes, and other elements. Segmentation was performed using various machine learning techniques including random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), and gradient-boosting machine (GBM), and 11 variables including geometry, color, intensity, and other road design features. MMS point cloud data for a 130-m section of a five-lane road near Minam Station in Busan, were used to evaluate the segmentation models; the average F1 scores of the models were 95.43% for RF, 92.1% for SVM, 91.05% for GBM, and 82.63% for KNN. The RF model showed the best segmentation performance, with F1 scores of 99.3%, 95.5%, 94.5%, 93.5%, and 90.1% for roads, sidewalks, curbs, medians, and lanes, respectively. The variable importance results of the RF model showed high mean decrease accuracy and mean decrease gini for XY dist. and Z dist. variables related to road design, respectively. Thus, variables related to road design contributed significantly to the segmentation of semantic information. The results of this study demonstrate the applicability of segmentation of MMS point cloud data based on machine learning, and will help to reduce the cost and effort associated with HD mapping.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.27 no.5
• /
• pp.113-119
• /
• 2023

Most infrastructure structures were completed during periods of economic growth. The number of infrastructure structures reaching their lifespan is increasing, and the proportion of old structures is gradually increasing. The functions and performance of these structures at the time of design may deteriorate and may even lead to safety accidents. To prevent this repercussion, accurate inspection and appropriate repair are requisite. To this end, demand is increasing for computer vision and deep learning technology to accurately detect even minute cracks. However, deep learning algorithms require a large number of training data. In particular, label images indicating the location of cracks in the image are required. To secure a large number of those label images, a lot of labor and time are consumed. To reduce these costs as well as increase detection accuracy, this study proposed a learning structure based on mean teacher method. This learning structure was trained on a dataset of 900 labeled image dataset and 3000 unlabeled image dataset. The crack detection network model was evaluated on over 300 labeled image dataset, and the detection accuracy recorded a mean intersection over union of 89.23% and an F1 score of 89.12%. Through this experiment, it was confirmed that detection performance was improved compared to supervised learning. It is expected that this proposed method will be used in the future to reduce the cost required to secure label images.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.6 s.38
• /
• pp.219-226
• /
• 2005

This paper describes the open API with integration of semantic web service with PARLAY X based open API in 4G mobile network. It can be expected that the intelligence such as the context-awareness, adaptation and personalization in 4G mobile network will be deployed. But the existing PARLAY based network lacks in considering context-awareness, adaptation and personalization. Therefore, the object of this paper is to support the architecture and the Application Programming Interface (API) of the network service for the context-awareness, adaptation and Personalization in 4G mobile network The open API is to provide users with the adaptive network service to the changing context constraints as well as detecting the changing context and user's Preference. For instance, the open API can Provide users with QoS in network according to the detected context and user's preference, after detecting the context such as location and speed and user's preference.

• Archives of design research
• /
• no.18
• /
• pp.153-162
• /
• 1996

underground space have many negative environmental clements which should be confirmed on location and line of action by the artificial signs. iccordingly, environmental graphics as visual language for popular signification-function have to be studied on sign theory of symbolic meanings. Ho\/ever, its usage has not only decorated by microscopiC appreciative eye but also lost its meanlllg as a visual language which was caused by the negligence of systematic management of the facility in charge Result of study, Visual environmental factors as a cause of behavioral attitude based on the study, which has been carefully considered as a communication of the visual language. Therefore, considering the underground environmental graphics as the: sign or the signification-function, It has to be studied on syntactic, semantic and pragmatic viewpoint. SpeCifically, to maKe the color and formation language a signification-function as a generalized connotation to the public, a distinctive classified Visual language must be applied.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.5
• /
• pp.1231-1242
• /
• 1997

This paper deals with an IP mukticast protocol in support of host mobility. Most of the previous multicast schemes have utilized an underlying logical strucuture of network topology,in order to provide a certain degree of order and predictability.On the other hand,mobility implies that a host location relaative to the rest of the net-work changes with time;the physical connectivity of the entire network is thus modified as move.In this case.some multicast datagrams nay not delivered properly,or may delivered twice or more,to a mobile host because the destinations will keep moving whlist datagrams are dekivered with different time delay.This paper first describes the relation between host mobility and multicast, by exploring the possible interactions,and presents a multicast scheme in support of Internet host mobility.A revised scheme is then proposed to adapt the multicast semantic and to optimize the communication overhead.

• Journal of KIISE:Software and Applications
• /
• v.35 no.7
• /
• pp.431-441
• /
• 2008

The academic information analysis service is including automatic recommendation of reviewers and analysis of researchers' accomplishments. The service of recommendation of reviewers should be processed in a transparent, fair and accountable way. When selecting reviewers, the following information must be considered: subject of project, reviewer's maj or, expertness of reviewer, relationship between applicant and reviewer. The analysis service of researchers' accomplishments is providing statistic information of researcher, institution and location based on accomplishments including book, article, patent, report and work of art. In order to support these services, we designed ontology for academic information, converted legacy data to RDF triples, expanded knowledge appropriate to services using OntoFrame. OntoFrame is service framework which includes ontology, reasoning engine, triple store. In our study, we propose the design methodology of ontology and service system for academic information based on OntoFrame. And then we explain the components of service system, processing steps of automatic recommendation of reviewers and analysis of researchers' accomplishments.