• Journal of Korea Multimedia Society
• /
• v.22 no.4
• /
• pp.432-442
• /
• 2019

Most current human action recognition methods based on deep learning methods. It is required, however, a very high computational cost. In this paper, we propose an action change detection method to reduce repetitive human action recognition tasks. In reality, simple actions are often repeated and it is time consuming process to apply high cost action recognition methods on repeated actions. The proposed method decides whether action has changed. The action recognition is executed only when it has detected action change. The action change detection process is as follows. First, extract the number of non-zero pixel from motion history image and generate one-dimensional time-series data. Second, detecting action change by comparison of difference between current time trend and local extremum of time-series data and threshold. Experiments on the proposed method achieved 89% balanced accuracy on action change data and 61% reduced action recognition repetition.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.35 no.10
• /
• pp.127-133
• /
• 2019

Damage to high-rise buildings caused by earthquakes is less frequency due to small distribution of high-rise buildings and low transmissibility of seismic motion to high-rise buildings. However, demand for high-rise buildings is increasing for development of construction technology and efficient land use. In addition, if high-rise buildings are constructed on soft ground such as landfill, transmissibility of seismic motion due to long-periodization of seismic waves is likely to increase. Thus, with development of technology, buildings are required to expand range of seismic design such as safety for long-period seismic waves. Therefore, in this study, dynamic experiments were performed to evaluate response characteristics of high-rise buildings according to period characteristics of seismicwaves and time history analysis was performed to verify them.

• Earthquakes and Structures
• /
• v.15 no.6
• /
• pp.605-616
• /
• 2018

In this paper, an extended Cloud analysis method is developed for seismic fragility assessment of existing highway bridges in the southeast Queensland region. This method extends the original Cloud analysis dataset by performing scaled Cloud analyses. The original and scaled Cloud datasets are then paired to generate seismic fragility curves. The seismic hazard in this region is critically reviewed, and the ground motion records are selected for the time-history analysis based on various record selection criteria. A parametric highway bridge model is developed in the OpenSees analysis software, and a sampling technique is employed to quantify the uncertainties of highway bridges ubiquitous in this region. Technical recommendations are also given for the seismic performance evaluation of highway bridges in such low-to-moderate seismic zones. Finally, a probabilistic fragility study is conducted by performing a total of 8000 time-history analyses and representative bridge fragility curves are generated. It is illustrated that the seismic fragility curves generated by the proposed extended Cloud analysis method are in close agreement with those which are obtained by the rigorous incremental dynamic analysis method. Also, it reveals that more than 50% of highway bridges existing in southeast Queensland will be damaged subject to a peak ground acceleration of 0.14 g.

• Structural Engineering and Mechanics
• /
• v.83 no.3
• /
• pp.401-413
• /
• 2022

The collapse of civil infrastructure due to natural disasters results in financial losses and many casualties. In particular, the recent increase in earthquake activities has highlighted on the importance of assessing the seismic performance and predicting the seismic risk of a structure. However, the nonlinear behavior of a structure and the uncertainty in ground motion complicate the accurate seismic response prediction of a structure. Artificial intelligence can overcome these limitations to reasonably predict the nonlinear behavior of structures. In this study, a deep learning-based algorithm was developed to estimate the time-history seismic response of bridge structures. The proposed deep neural network was trained using structural and ground motion parameters. The performance of the seismic response prediction algorithm showed the similar phase and magnitude to those of the time-history analysis in a single-degree-of-freedom system that exhibits nonlinear behavior as a main structural element. Then, the proposed algorithm was expanded to predict the seismic response and fragility prediction of a bridge system. The proposed deep neural network reasonably predicted the nonlinear seismic behavior of piers and bearings for approximately 93% and 87% of the test dataset, respectively. The results of the study also demonstrated that the proposed algorithm can be utilized to assess the seismic fragility of bridge components and system.

• International Journal of Ocean System Engineering
• /
• v.1 no.4
• /
• pp.211-221
• /
• 2011

Tugboats are widely used near harbors to assist with various operations such as the berthing and deberthing of very large vessels and the towing of barges. Capsizing accidents involving tugboats occasionally take place when the tugboat makes rapid turns in harsh weather conditions. When there is little evidence suggesting how the accident occurred and when the crew members are missing, it is necessary to predict the time history of the towing vessel’s attitude and trajectory from its departure point to when and where it capsized, depending on various input parameters using a numerical simulation. In this paper, the dynamics of a tugboat and a towed barge in conjunction with the external force and moment were established, and the possible input parameters and operational scenarios which might influence the large roll motion of the tugboat were identified. As a result of analyzing the simulated time history of the excessive roll motion of the tugboat, it was found that roll motion can take place when the tugboat is situated on the crest of a wave and when it is pulled by a towed barge through a towing line. The main cause of the accident would be the parameters that primarily influence such situations. These are the wave parameters, course changing scenario, and the amount of tension.

• Journal of Korean Association for Spatial Structures
• /
• v.18 no.3
• /
• pp.105-116
• /
• 2018

The objective of this study is to investigate the response reducing effect of a seismic isolation system installed between 300m dome and supports under both horizontal and vertical seismic ground motion. The time history analysis is performed to investigate the dynamic behavior of single layer lattice domes with and without a lead rubber bearing seismic isolation system. In order to ensure the seismic performance of lattice domes against strong earthquakes, it is important to investigate the mechanical characteristics of dynamic response. Horizontal and vertical seismic ground motions cause a large asymmetric vertical response of large span domes. One of the most effective methods to reduce the dynamic response is to install a seismic isolation system for observing seismic ground motion at the base of the dome. This paper discusses the dynamic response characteristics of 300m single layer lattice domes supported on a lead rubber seismic isolation device under horizontal and vertical seismic ground motions.

• Earthquakes and Structures
• /
• v.16 no.4
• /
• pp.487-499
• /
• 2019

Energy-based seismic design of structures has gradually become prominent in today's structural engineering investigations because of being more rational and reliable when it is compared to traditional force-based and displacement-based methods. Energy-based approaches have widely taken place in many previous studies and investigations and undoubtedly, they are going to play more important role in future seismic design codes, too. This paper aims to compute the maximum earthquake energy input to elastic single-degree-of-freedom (SDOF) systems for selected real ground motion records. A data set containing 100 real ground motion records which have the same site soil profiles has been selected from Pacific Earthquake Research (PEER) database. Response time history (RTH) analyses have been conducted for elastic SDOF systems having a constant damping ratio and natural periods of 0.1 s to 3.0 s. Totally 3000 RTH analyses have been performed and the maximum mass normalized earthquake input energy values for all records have been computed. Previous researchers' approaches have been compared to the results of RTH analyses and an approach which considers the pseudo-spectral velocity with Arias Intensity has been proposed. Graphs of the maximum earthquake input energy versus the maximum pseudo-spectral velocity have been obtained. The results show that there is a good agreement between the maximum input energy demands of RTH analysis and the other approaches and the maximum earthquake input energy is a relatively stable response parameter to be used for further seismic design and evaluations.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.2A
• /
• pp.273-284
• /
• 2006

In this study, After considering the general characteristic of Near Fault Ground Motion, the inelastic response spectrum is made to evaluate using the change of ductility and yield stiffness coefficient according to the inelastic behavior of structures which couldn't be examined through the elastic response spectrum. It is conducted to the elastic and inelastic time history analysis about the long period structure which could reflect the characteristic of Near Fault Ground Motion with the best and it is also examined the aspect of response distribution about the input data. Moreover, the response characteristic of structure is analyzed by investigating the plastic hinge for the purpose of grasp about the inelastic behavior of structure.

• The Journal of the Korea Contents Association
• /
• v.17 no.6
• /
• pp.632-642
• /
• 2017

The purpose of this study was to determine influence factors for predicting postoperative nausea and vomiting(PONV) for postoperative 24hr. Participants were 332 general surgery(GS), neurosurgery(NS), orthopedic surgery(OS), obstetrics and gynecology(OBGY), otorhinolaryngology (ENT) surgical patients at one hospital in G city. Data were collected from December 2014 to December 2015, and the data were analyzed by t-test, $x^2$-test and logistic regression analysis. Multivariate analysis revealed that non smoking (p=.011), history of motion sickness (p=.020), history of PONV (p=.001), post operative pain (p=.001) were influence factors for PONV. Preoperative intensive management of PONV is necessary if there is a history of non smoker, history of motion sickness and history of PONV before surgery, and postoperative pain should be actively intervened.

• Earthquakes and Structures
• /
• v.11 no.6
• /
• pp.1027-1041
• /
• 2016

Different incident angles of ground motions have been considered to evaluate the relationship between floor rotation and torsional irregularity coefficient. The issues specifically addressed are (1) variability in torsional irregularity coefficient and floor rotations with varying incident angles of ground motion (2) contradictory relationship between floor rotation and torsional irregularity coefficient. To explore the stated issues, an evaluation based on relative variation in seismic response quantities of linear asymmetric structure under the influence of horizontal bi-directional excitation with varying seismic orientations has been carried out using response history analysis. Several typical earthquake records are applied to the structure to demonstrate the relative variations of floor rotation and torsional irregularity coefficient for different seismic orientations. It is demonstrated that (1) Torsional irregularity coefficient (TIC) increases as the story number decreases when the ground motion is considered along reference axes of the structure. For incident angles other than structure's reference axes, TIC either decreases as the story number decreases or there is no specific trend for TIC. Floor rotation increases in proportion to the story number when the ground motion is considered along reference axes of structure. For incident angles other than structure's reference axes, floor rotation either decreases as the story number increases or there is no specific trend for floor rotation and (2) TIC and floor rotation seems to be approximately inversely proportional to each other when the ground motion is considered along reference axes of the structure. For incident angles other than structure's reference axes, the relationship can even become directly proportional instead of inversely proportional.