• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.3
• /
• pp.108-123
• /
• 2007

As H.264/AVC standard has proven to be a key technology of multimedia application, many researches to improve H.264/AVC standard are actively conducted. Those researches are conducted in various ways such as algorithm analysis and improvement or structure enhancement for reducing bottlenecks of performance. Even though targets and directions of those studies are not the same, performance of H.264/AVC standard is commonly analyzed in the early phase. In analysis phase, potential problems with H.264/AVC standard are identified and the most critical problem which has serious effects on performance is determined. Therefore, analysis phase is one of the important steps to decide overall directions and targets of the research. This research proposes a mathematical model which can be used in the early performance analysis phase to estimate performance in conducting research of improving the performance of H.264/AVC Baseline Profile decoder. The proposed model is designed by considering many variables of H.264/AVC decoder operation so that it is easy to predict its performance according to changes in each element.

• Smart Structures and Systems
• /
• v.24 no.2
• /
• pp.243-256
• /
• 2019

Determination of the most meaningful structural modes and gaining insight into how these modes evolve are important issues for long-term structural health monitoring of the long-span bridges. To address this issue, modal parameters identified throughout the life of the bridge need to be compared and linked with each other, which is the process of mode tracking. The modal frequencies for a long-span bridge are typically closely-spaced, sensitive to the environment (e.g., temperature, wind, traffic, etc.), which makes the automated tracking of modal parameters a difficult process, often requiring human intervention. Machine learning methods are well-suited for uncovering complex underlying relationships between processes and thus have the potential to realize accurate and automated modal tracking. In this study, Gaussian mixture model (GMM), a popular unsupervised machine learning method, is employed to automatically determine and update baseline modal properties from the identified unlabeled modal parameters. On this foundation, a new mode tracking method is proposed for automated mode tracking for long-span bridges. Firstly, a numerical example for a three-degree-of-freedom system is employed to validate the feasibility of using GMM to automatically determine the baseline modal properties. Subsequently, the field monitoring data of a long-span bridge are utilized to illustrate the practical usage of GMM for automated determination of the baseline list. Finally, the continuously monitoring bridge acceleration data during strong typhoon events are employed to validate the reliability of proposed method in tracking the changing modal parameters. Results show that the proposed method can automatically track the modal parameters in disastrous scenarios and provide valuable references for condition assessment of the bridge structure.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2005.05a
• /
• pp.684-691
• /
• 2005

본 연구는 전달함수 합성법을 이용하여 보강후 구조물의 진동특성을 예측하는 방법이다. 이 방법은 보강후 구조물의 진동특성을 보강전 구조물의 전달함수를 이용하여 예측하는 기술로, 이에 관한 이론은 많이 알려져 있다. 하지만 실제 실험으로 전달함수를 계측할 경우 회전자유도에 대한 전달함수를 계측하기가 어렵다. 따라서 병진자 유도만으로 전달함수 합성법을 적용할 경우 발생하는 고유진동수 추정 오차를, 전체 자유도를 이용한 경우와 간단한 구조물의 수치해석을 통해서 비교해 보고자 한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.4
• /
• pp.1704-1720
• /
• 2020

Recognizing food from photographs presents many applications for machine learning, computer vision and dietetics, etc. Recent progress of deep learning techniques accelerates the recognition of food in a great scale. We build a hierarchical structure composed of deep CNN to recognize and classify food from photographs. We build a dataset for Korean food of 18 classes, which are further categorized in 4 major classes. Our hierarchical recognizer classifies foods into four major classes in the first step. Each food in the major classes is further classified into the exact class in the second step. We employ DenseNet structure for the baseline of our recognizer. The hierarchical structure provides higher accuracy and F1 score than those from the single-structured recognizer.

• Speech Sciences
• /
• v.15 no.3
• /
• pp.117-131
• /
• 2008

The purpose of this study was to determine the effects of whole language approach on the development of language abilities and story structure concepts for hearing impaired children. For this end, two research questions have been established. First, what is the effect of whole language program using story books on hearing impaired children’s language abilities? Second, what is the effect of whole language program using story books on hearing impaired children's story structure concept? Three subjects participated in the study. Each subject was scheduled for a 40-minute session two times a week. Subjects received 36 sessions of use animation activities for 3 months. The study used a multiple baseline across the subjects. The followings were the findings of this study. First, the whole language program using story books improved hearing impaired children's language abilities. Second, the whole language program using story books improved hearing impaired children's story structure concept.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.11
• /
• pp.2903-2923
• /
• 2023

Knowledge recommendation is a type of recommendation system that recommends knowledge content to users in order to satisfy their needs. Although using graph neural networks to extract data features is an effective method for solving the recommendation problem, there is information loss when modeling real-world problems because an edge in a graph structure can only be associated with two nodes. Because one super-edge in the hypergraph structure can be connected with several nodes and the effectiveness of knowledge graph for knowledge expression, a dual-channel hypergraph convolutional neural network model (DCHC) based on hypergraph structure and knowledge graph is proposed. The model divides user data and knowledge data into user subhypergraph and knowledge subhypergraph, respectively, and extracts user data features by dual-channel hypergraph convolution and knowledge data features by combining with knowledge graph technology, and finally generates recommendation results based on the obtained user embedding and knowledge embedding. The performance of DCHC model is higher than the comparative model under AUC and F1 evaluation indicators, comparative experiments with the baseline also demonstrate the validity of DCHC model.

• Journal of The Korean Astronomical Society
• /
• v.38 no.2
• /
• pp.261-266
• /
• 2005

This paper reviews the progress in the VLBI (Very Long Baseline Interferometry) studies of Sgr A$\ast$, the best known supermassive black hole candidates with a dark mass concentration of $4 {\times} 10^6 M_{\bigodot}$ at the center of the Milky Way. The emphasis is on the importance of the millimeter and sub-millimeter VLBI observations in the detection of Sgr A$\ast$'s intrinsic structure and search for the structural variation.

• Proceedings of the KSR Conference
• /
• 2007.11a
• /
• pp.913-916
• /
• 2007

In this paper, the possibility of time reversal phenomenon was investigated in damage detection of structure. In conventional lamb wave techniques, damage is identified by comparing the measured data (baseline signals) and the current data. But this method can lead to high false signal in the intact condition of structures due to environmental conditions of the structures. So in this studying, we investigate the possibility of damage detection in the aluminum plate using the time reversal phenomenon of lamb waves.

• Korean System Dynamics Review
• /
• v.7 no.1
• /
• pp.213-228
• /
• 2006

This paper reports the results of a study designed to understand and facilitate disaster mitigation for communities located in low frequency/high magnitude earthquake zones. The study is based on a small town located near the New Madrid Fault Zone and is therefore at significant earthquake risk. A system dynamics model describes the variables and policies governing the distribution of building safety over time. Data from this town is used to establish a 25-year baseline. Simulations are run to demonstrate the consequences of different building policies.

• Smart Structures and Systems
• /
• v.15 no.3
• /
• pp.787-806
• /
• 2015

Due to corrosion, a large number of belt conveyors support structure in industrial plants have deteriorated. Severe corrosion may result in collapse of the structures. Therefore, practical and effective structural assessment techniques are needed. In this paper, damage identification methods based on two specific local vibration modes, named periodic and isolated local vibration modes, are proposed. The identification methods utilize the facts that support structures have many identical members repeated along the belt conveyor and there exist some local modes within a small frequency range where vibrations of these identical members are much larger than those of the other members. When one of these identical members is damaged, this member no longer vibrates in those modes. Instead, the member vibrates alone in an isolated mode with a lower frequency. A damage identification method based on frequencies comparison of these vibration modes and another method based on amplitude comparison of the periodic local vibration mode are explained. These methods do not require the baseline measurement records of undamaged structure. The methods is capable of detecting multiple damages simultaneously. The applicability of the methods is experimentally validated with a laboratory model and a real belt-conveyor support structure.