• Genomics & Informatics
• /
• v.19 no.3
• /
• pp.22.1-22.5
• /
• 2021

Automatic document classification for highly interrelated classes is a demanding task that becomes more challenging when there is little labeled data for training. Such is the case of the coronavirus disease 2019 (COVID-19) clinical repository-a repository of classified and translated academic articles related to COVID-19 and relevant to the clinical practice-where a 3-way classification scheme is being applied to COVID-19 literature. During the 7th Biomedical Linked Annotation Hackathon (BLAH7) hackathon, we performed experiments to explore the use of named-entity-recognition (NER) to improve the classification. We processed the literature with OntoGene's Biomedical Entity Recogniser (OGER) and used the resulting identified Named Entities (NE) and their links to major biological databases as extra input features for the classifier. We compared the results with a baseline model without the OGER extracted features. In these proof-of-concept experiments, we observed a clear gain on COVID-19 literature classification. In particular, NE's origin was useful to classify document types and NE's type for clinical specialties. Due to the limitations of the small dataset, we can only conclude that our results suggests that NER would benefit this classification task. In order to accurately estimate this benefit, further experiments with a larger dataset would be needed.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.213-216
• /
• 2001

본 논문은 DCT, FFT 및 Wavelet 등과 같은 주파수 영역에 의한 Watermarking 기법에서 정확한 Watermark를 검출하기 위하여 제안한 알고리즘이다. 본 논문에서 사용한 알고리즘은 기존의 왜곡된 영상에서 원 영상의 주파수 변환 값을 제거하여 왜곡된 상태의 영상 값에서 삽입된 Watermark를 검출하는 알고리즘에서 벗어나 왜곡된 값을 완전히 제거한 상태에서 영상의 Watermark를 검출함으로써 보다 완벽한 Watermark를 검출 할 수 있는 향상된 알고리즘을 제안하였다.

• Physical Therapy Korea
• /
• v.19 no.3
• /
• pp.1-10
• /
• 2012

The aim of this study is to compare measurements of abdominal muscle thickness using ultrasonography imaging (USI) to those using a special transducer head device, during five different trunk stabilization exercises, ultimately to determine which exercise led to the greatest muscle thickness. Thirty eight healthy subjects participated in this cross-sectional study. The five types of trunk stabilization - i.e., a sit-up on the supine, an upper and lower extremity raise with quadruped on the prone, a leg raise in sitting on the ball, trunk rolling on the ball, and balance using sling on the prone position - were each performed with an abdominal draw. The thickness of the abdominal muscle - including the transverse abdominal (TrA), internal oblique (IO), and external oblique (EO) - was measured by USI with a special transducer head device, at rest and then at contraction in each position. Data were analyzed using one-way repeated ANOVA with the level of significance set at ${\alpha}$=.05. The results were as follows: 1) the TrA thickness was statistically significant (p<.05), whereas the IO and EO thicknesses were not (p>.05); 2) among the five types of trunk stabilization, TrA thickness significantly increased with the balance using a sling in the prone position, (p<.05), whereas no significant difference was noted for the four types of trunk stabilization (p>.05); 3) reliability data showed that there was a high degree of consistency among the measurements taken using the special transducer head device (ICC=.92). In conclusion, the balance using a sling in the prone position was more effective than any of the four other types of trunk stabilization in increasing TrA thickness in healthy subjects.

• Journal of Broadcast Engineering
• /
• v.20 no.3
• /
• pp.430-439
• /
• 2015

The demand of violence detection techniques using a video analysis to help prevent crimes is increasing recently. Many researchers have studied vision based behavior recognition but, violent behavior analysis techniques usually focus on violent scenes in television and movie content. Many methods previously published usually used both a color(e.g., skin and blood) and motion information for detecting violent scenes because violences usually involve blood scenes in movies. However, color information (e.g., blood scenes) may not be useful cues for violence detection in surveillance videos, because they are rarely taken in real world situations. In this paper, we propose a method of violent behavior detection in surveillance videos using motion vectors such as flow vector magnitudes and changes in direction except the color information. In order to evaluate the proposed algorithm, we test both USI dataset and various real world surveillance videos from YouTube.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.7 no.4
• /
• pp.31-37
• /
• 2003

This paper presents the methodologies for seismic retrofitting of cabinet equipment which can be employed to resolve the USI A-46 problem related to seismic qualification of old nuclear power plant. To obtain accurate dynamic characteristics of a cabinet structure, three types of structural modeling are introduced and the their free vibration modes are compared. Three types of seismic retrofitting scenarios, such as 1) the installation of bracing, 2) installation of damper, 3) installation of tuned mass damper(TMD), are established and evaluated for the decrease of ICRS(In Cabinet Reponse Spectrum). In the cases of 1) ＆ 2), since the retrofitted structures show larger ICRS than that of the original structure, the careful considerations are need in the application of these methods. Though the installation of TMD shows the best retrofitting result, the construction of analysis model that indicate the accurate vibration modes of real structure is estimated the essential step of this retrofitting method.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2009.01a
• /
• pp.113-116
• /
• 2009

현재 우리의 7차 교육과정에서 교육의 보조 도구로써 교실에서 사용될 수 있는 것은 별로 많지 않다. 그래서 AvaLab은 이 시점에서 좋은 도구가 될 수 있다. AvaLab은 컴퓨터에 익숙하지 않은 학생을 위한 GUI 프로그래밍 도구이며, 현재의 교육 과정에서 필요로 하는 창의성을 키우기 위한 한 도구로써 교실에서 사용될 수 있다. 사용하려면, 먼저 학습자는 프로그래밍 아이디어를 생각하고, 아이콘 기반의 AvaLab을 통하여 프로그래밍한다. 프로그래밍 후에 학습자는 로봇의 결과를 생각하고 토론할 수 있다. 만약 그 결과가 정확화지 않다면, 로봇은 피드백을 보일 젓이다. 그러므로, AvaLab은 초보자를 위한 도구이며, 7차 교육과정에 적합한 도구이다. AvaLab의 특징 중 하나는 아이콘 기반의 흐름도 모델로 구성되어 있다는 것과 학습자는 입력과 출력 센스를 갖는 로봇을 이용해 실습할 수 있다는 것이다.

• Proceedings of the Korea Contents Association Conference
• /
• 2003.11a
• /
• pp.218-222
• /
• 2003

A technology that recognize the car license plate have accomplished a lot of developments for latest several years. Key technology for correct recognition is correct abstraction of plate area. Existent studies have used horizontal/vertical edge, some geometrical characteristics of license plate, and the color information. But, in case of extracting a plate using above characteristics, correct extraction of a license plate inclined by sight which see license plate is difficult. Therefore, this paper is propose new method that correctly extract license plate using the color information and linear regression method.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2003.04a
• /
• pp.374-381
• /
• 2003

In this study, the reduction schemes of ICRS(In-Cabinet Response Spectra), the main part of seismic safety qualification of old nuclear power plant(USI A-46 problem), are studied. To obtain accurate dynamic characteristics of cabinet structure, the cabinet structure modeled by frame model and its eigen analysis is performed. The three types of seismic retrofitting scenarios, such as 1) the installation of bracing, 2) installation of damper, 3) installation of tuned mass damper(TMD), are established and evaluated for the decreasing of ICRS. In the cases of 1) ＆ 2), since, the retrofitted structures show larger ICRS than that of original structure, the careful considerations are need in the application of these methods. Though, the case of TMD show the best retrofitting result, the tuning between the real structure and analysis model is estimated the essential step of retrofitting.

• Economic and Environmental Geology
• /
• v.11 no.4
• /
• pp.183-186
• /
• 1978

Chemical analyses by electron microprobe indicate that the well-crystallized botryoidal coffinite with radial fibrous structure from the Woodrow Mine, New Mexico, U.S.A. has the formula ($U_{0.85}Ca_{0.14}$) $(Si_{0.86}P_{0.11}Al_{0.06}V_{0.02})O_2(OH)_4$, or ideally (U, Ca) (Si, P)$O_2(OH)_4$. It can be more simply designated as ideal formula $USiO_2(OH)_4$. This new formula is proposed for the coffinite.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.33 no.2
• /
• pp.232-240
• /
• 2013

This paper presents recent advances in damage visualization algorithms of laser generated ultrasonic propagation imaging(UPI) system. An effective damage evaluation method is required to extract correct information from raw data to properly characterize anomalies present in structure. A temporal-reference free imaging system provides easy and rapid defect inspection capability with less computational complexity. In this paper a number of methods such as ultrasonic wave propagation imaging(UWPI), anomalous wave propagation imaging(AWPI), ultrasonic spectral imaging(USI), wavelet ultrasonic propagation imaging(WUPI), variable time window amplitude mapping(VTWAM), time point adjustment(TPA), time of flight and amplitude mapping(ToF&Amp) and ultrasonic wavenumber imaging(UWI) are discussed with instances of successful implementation on various structures.