• International journal of advanced smart convergence
• /
• v.4 no.2
• /
• pp.20-28
• /
• 2015

In this paper, we propose an effective tracking algorithm with an appearance model based on features extracted from a video frame with posture variation and camera view point adaptation by employing the non-adaptive random projections that preserve the structure of the image feature space of objects. The existing online tracking algorithms update models with features from recent video frames and the numerous issues remain to be addressed despite on the improvement in tracking. The data-dependent adaptive appearance models often encounter the drift problems because the online algorithms does not get the required amount of data for online learning. So, we propose an effective tracking algorithm with an appearance model based on features extracted from a video frame.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.25 no.1
• /
• pp.28-36
• /
• 2013

The aim of this study is to evaluate the performance of the GHP system with 45 cast-in-place energy piles(CEP) for a commercial building. In order to demonstrate the feasibility of a sustainable performance of the system, transient simulations were conducted over 1-year and 20-year periods, respectively. The 1-year simulation results showed that the maximum and minimum temperatures of brine returning from the CEPs were $23.91^{\circ}C$ and $6.66^{\circ}C$, which were in a range of design target temperatures. In addition, after 20 years' operation, these returning temperatures decreased to $21.24^{\circ}C$ and $3.68^{\circ}C$, and finally reached to stable state. Annual average extraction heat of cast-in-place energy piles was 94.3 MWh and injection heat was 65.7 MWh from the 20 years of simulation results. Finally, it is expected this GHP system can operate with average heating SPF of more than 3.45 for long-term.

• Korean Journal of Bronchoesophagology
• /
• v.10 no.1 s.19
• /
• pp.55-57
• /
• 2004

On occasion there were reports of foreign body of paranasal sinuses. Most common site is the maxillary sinus. But it is very rare to experience a long-term foreign body in maxillary sinus. There are two types of maxillary foreign bodies according to etiology, one is caused by various traumatic accidents, and the other is iatrogenic cause which mainly retaining gauze or medical instruments after sinus operation or teeth extraction. We experienced an interesting case of over fifty yews resided metal foreign body in maxillary sinus caused sinusitis, and report with a brief literature review.

• Journal of Oral Medicine and Pain
• /
• v.46 no.3
• /
• pp.88-92
• /
• 2021

Acute osteomyelitis caused by Klebsiella pneumoniae is rare in the oral and maxillofacial region. Klebsiella pneumoniae is a Gram-negative bacillus and the normal flora of the human body, but it can cause pneumonia, urinary tract infection, meningitis, and osteomyelitis in patient with compromised immune systems. These infections are mainly caused by nosocomial infection. Microbacterial osteomyelitis was developed by clinical cause such as tooth extraction, fracture, and surgical history, which requires long-term antibiotic administration and surgical treatment. This report describes that a 56-year-old male patient with acute osteomyelitis caused by Klebsiella pneumoniae infection after implant placement was treated with intravenous administration of ertapenem without open surgery treatment. Through this case, we report that antibiotic susceptibility test is essential for the treatment of acute osteomyelitis caused by a bacterial infection resistant to empirical antibiotics, and early administration of appropriate antibiotics can reduce the possibility of extensive bone destruction or additional open surgery.

• Nuclear Engineering and Technology
• /
• v.51 no.4
• /
• pp.1041-1046
• /
• 2019

In terms of long-term safety for radioactive waste disposal, the anionic iodide (I-129) with a long half-life ($1.6{\times}10^6yr$) is of a critical importance because this radionuclide migrates in geological media with limited interactions. Various studies have been performed to retard the iodide migration. Recently, some minerals that are likely generated from waste container corrosion, have been suggested to have a considerable chemical interaction with iodide. In this study, chalcocite and mackinawite were selected as candidate minerals for underground corrosion materials, and an iodide sorption experiment were carried out. The experiment was performed under anoxic and alkaline conditions and the pH effects on the iodide sorption were investigated in the range of pH 8 to 12. The results showed that both minerals demonstrated a noticeable sorption capacity on iodide, and the distribution coefficient ($K_d$) decreased as the pH increased in the experimental condition. In addition, when the alkalinity increased higher than a pH of 12, the sorption capacity of both minerals decreased dramatically, likely due to the competition of hydroxy ions with the iodide. This result confirmed that chalcocite was an especially good sorbing media for iodide under alkaline conditions with a pH value of less than 12.

• Genomics & Informatics
• /
• v.17 no.2
• /
• pp.20.1-20.5
• /
• 2019

Entity normalization, or entity linking in the general domain, is an information extraction task that aims to annotate/bind multiple words/expressions in raw text with semantic references, such as concepts of an ontology. An ontology consists minimally of a formally organized vocabulary or hierarchy of terms, which captures knowledge of a domain. Presently, machine-learning methods, often coupled with distributional representations, achieve good performance. However, these require large training datasets, which are not always available, especially for tasks in specialized domains. CONTES (CONcept-TErm System) is a supervised method that addresses entity normalization with ontology concepts using small training datasets. CONTES has some limitations, such as it does not scale well with very large ontologies, it tends to overgeneralize predictions, and it lacks valid representations for the out-of-vocabulary words. Here, we propose to assess different methods to reduce the dimensionality in the representation of the ontology. We also propose to calibrate parameters in order to make the predictions more accurate, and to address the problem of out-of-vocabulary words, with a specific method.

• The Korean Journal of Food & Health Convergence
• /
• v.5 no.3
• /
• pp.7-20
• /
• 2019

The use of caffeine as a psychoactive stimulant in tea has been observed to have serious negative effects in humans' systems such as respiratory, nervous, cardiovascular, renal and skeletal systems. This study was carried out to assess the levels of caffeine in 10 different tea brands available in local market in Dutse, Jigawa State, Nigeria. Quantitative analysis of caffeine was performed by a simple and fast UV-Vis spectrophotometric methods using different solvents for extraction. The caffeine content in all the tea samples analyzed in this study were below the maximum allowable limits set by the USFDA. Tea have been associated with adverse health effects and the claims made by manufacturers about the benefits of tea do not highlight risks associated with excessive consumption of a combination of the ingredients contained in tea. Long term effects of tea consumption of children and young people have not been adequately studied. Therefore, it is recommended that further research be carried out on the adverse effects of energy drinks on children. Research is also needed to be done on the effects of the combination of ingredients on health and excessive consumption of those ingredients to children and adolescents. People need to be educated and given proper awareness on the health risks associated with caffeine containing beverages.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4664-4681
• /
• 2020

Modulation recognition (MR) plays a key role in cognitive radar, cognitive radio, and some other civilian and military fields. While existing methods can identify the signal modulation type by extracting the signal characteristics, the quality of feature extraction has a serious impact on the recognition results. In this paper, an end-to-end MR method based on long short-term memory (LSTM) and the gated recurrent unit (GRU) is put forward, which can directly predict the modulation type from a sampled signal. Additionally, the sliding window method is applied to fast-changing mixed-modulation signals for which the signal modulation type changes over time. The recognition accuracy on training datasets in different SNR ranges and the proportion of each modulation method in misclassified samples are analyzed, and it is found to be reasonable to select the evenly-distributed and full range of SNR data as the training data. With the improvement of the SNR, the recognition accuracy increases rapidly. When the length of the training dataset increases, the neural network recognition effect is better. The loss function value of the neural network decreases with the increase of the training dataset length, and then tends to be stable. Moreover, when the fast-changing period is less than 20ms, the error rate is as high as 50%. As the fast-changing period is increased to 30ms, the error rates of the GRU and LSTM neural networks are less than 5%.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.44 no.1
• /
• pp.17-25
• /
• 2021

Ammunition Demilitarization facility (ADF) should be set up the feasible goals and continue to operate, taking into account non-profit characteristics. However, due to the lack of performance measurement methods in ADF, which are essential to national policy at a significant cost each year, the reliability of the evaluation results can be insufficient. In this paper, the Balanced Score Card (BSC) method was applied that could be evaluated to reflect the financial and non-financial features. The relevant literature research and army regulations reflected the results of various interviews of the expert group. The extraction of success performance area in ADF was confirmed using the BSC method and the Decision Variable (DV) candidate was created to use regression for selecting the DV. Additionally, the key performance indicator was presented by verification the feasibility of content by conducting the survey of experts. The implications of this paper are as follows. First, the proposed BSC model was found to be suitable for practical use in ADF reflecting the non-profit characteristics. Second, accurate evaluation of ADF can contribute to long-term development of ADF. Finally, it can be applied to the management process of the other military sector, so it can be expected to play a role in providing basic data and spreading it to other areas.