• 한국작물학회:학술대회논문집
• /
• 한국작물학회 2017년도 9th Asian Crop Science Association conference
• /
• pp.154-154
• /
• 2017

In accordance with the opening of the Korean rice market, this study was focused on establishment of database for discriminating the Korean rice varieties and imported brand rices using DNA markers. In this study, the SNP markers were developed using single nucleotide polymorphisms between the reference sequences of japonica and them of 40 brand rices which collected in Australia, China, Thailand, United States and Vietnam. The developed SNP markers were screened to a total of 360 rices including 320 Korean rice varieties and 40 imported brand rices. We selected polymorphic markers among Korean bred rive varieties and imported brand rices. The selected markers were classified into 3 grades. The markers of A grade produced DNA band in 360 rices of 30~40%, B grades produced in 40~60%, and C grades produced bands over 60% rices. First, we tried to set-up the discriminating system using the minimum SNP markers of A grade. Especially, a set of sixteen SNP markers could identify among Korean bred rice varieties and imported brand rices. Additionally, some SNP markers like NSb for Pib gene, JJ80-T for Pi5 and YL155/YL87 for Pita which linked to resistance genes to blast were used to fingerprinting system. These markers were set-up as multiplex set for enhancing the identification efficiency among rice varieties. Finally, the selected SNP markers would be used to the fluidigm assay to construct the database for elaborate discrimination of rice varieties.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 하계학술대회 논문집 B
• /
• pp.1406-1408
• /
• 2004

Recently, the load increasement and regenerative system of electrified railway system make a difficult to distinguish between the load current and fault current and it need a new intelligent protection system. In this paper, we proposed intelligent algorithm for discriminating normal and abnormal situation instead of the system being operated abnormally.

• 한국감성과학회:학술대회논문집
• /
• 한국감성과학회 1998년도 추계학술발표 논문집
• /
• pp.230-234
• /
• 1998

This study developed a test with a high test-retest reliability consisting of 18 items by selecting from Korean adjectives describing tactile sensations. Principal components were extracted by three repeated measurements of 19 different fabrics for underwear. Adjectives such as 'smooth' 'soft' 'sexy' 'comfortable' and 'cozy' were most discriminating between preferred and non-preferred fabrics. The fabrics were analyzed by Kawabata Evaluation System in 16 mechanical properties, then these properties were analyzed by regression analyses as to their relations with human sensation and sensibility. Among the fabrics, the most preferred were made of polyester whereas the least preferred were made of gauze and cotton. Mechanical properties were identified that were related to the most discriminating adjectives such as 'smooth', 'soft', 'sexy,' and 'cozy'.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1998년도 추계종합학술대회 논문집
• /
• pp.617-620
• /
• 1998

In this paper we realized non-contact sheet resistance measurement system using eddy current. Proposed system is designed to meet the requirements which is necessary when dealing with conducting thin films on large area LCD panel. With several metals we could get lift-off curves which has the same trend as in principal. Especially in the region of high conductivity this system has more discriminating ability than 4-point probe system.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 1987년도 학술발표회 논문집
• /
• pp.89-92
• /
• 1987

In a template-matchig-based speech recognition syste, excessive weight given to perceptually unimportant spectral variations is undesirable for discriminating among acoustically similar words. By introducing a simple threshold-type nonlinearity applied to the distance metric, the word recognition performance can be improved for a vocabulary with similar sounding words, without modifying the system structure.

• 임상간호연구
• /
• 제19권3호
• /
• pp.357-368
• /
• 2013

Purpose: The purpose of this study was to investigate risk factors related to delirium and to develop screening model on delirium occurrence in MICU (Medical Intensive Care Unit) patients. Methods: For developing a preliminary tool for delirium, the data of 166 patients were collected and analyzed. In order to estimate the accuracy and discriminating power for the developed screening model, 98 patients were enrolled. The data used in this study were collected by EMR (Electronic Medical Record) review from January to September in 2012. The collected data were analyzed using SPSS/PC Win 18.0 program. Results: Screening model on delirium in MICU patients was developed using the results of logistic regression. The total score of screening model was 24 point and measuring point was 10 point. When the measuring point is over 10 point, it means that the risk of delirium occurrence is high. The discriminating power and the validity of screening model showed AUC .908 (p <.001) and .935 (p <.001) respectively. This result showed that the screening model on delirium which developed in this study was an appropriate model for screening the delirium risk group in MICU. The sensitivity of the screening model was 83%, specificity 89% and accuracy 84%. Conclusion: The developed screening model on delirium occurrence in MICU should be combined with EMR for screening and preventing delirium in a high risk group.

• 동의생리병리학회지
• /
• 제24권5호
• /
• pp.883-891
• /
• 2010

The purpose of this study was to investigate the relation between each pattern identification such as Dampness-Phlegm, Fire-Heat, Deficiency of Qi, Deficiency of Yin and Blood Stagnation and blood lipid level in acute stroke patients. This study was done over patients hospitalized in 13 Oriental Medical Hospitals in the period of November 2006 to Jun 2009. Patients had been interviewed by residents and medical specialists who studied standard operation procedures in Fundamental Study for Syndrome of Oriental Medicine for Stroke. Study subjects consisted of group A that was distributed to specified pattern identification by medical specialist and discriminating program, group B that was distributed to specified pattern identification by medical specialist or discriminating program, and the control group that wasn't distributed to specified pattern identification by medical specialist and discriminating program. For the purpose of obtaining suitable result, we analyzed blood lipid level of each group by univariate analysis. In this study, there was not statistically significant relation between most of each pattern identification and blood lipid level except relation between Fire-Heat pattern group B and decreasing HDL cholesterol. Based on these results, it is suggested that Dampness-Phlegm would not be the independent predictors of hyperlipidemia unlike other studies that were presented in. More prospective studies between Fire-Heat and decreasing HDL cholesterol are to be done with more clinical data.

• 정보보호학회논문지
• /
• 제31권3호
• /
• pp.387-400
• /
• 2021

클라우드 환경의 발전에 따른 스트리밍 플랫폼과 콘텐츠의 증가로 합법적인 콘텐츠뿐만 아니라 불법 복제된 콘텐츠 또한 빠른 속도로 증가하고 있다. 따라서 다양한 콘텐츠에 대한 저작권 침해 여부를 판별할 수 있는 기술의 개발이 요구된다. 한국저작권보호원에서는 AI를 활용한 영상 콘텐츠 실증시스템을 운영하고 있지만, 해상도 변화와 같은 왜곡에 대해 한계점을 가지고 있다. 본 논문에서는 스켈레톤 정보를 활용하여 불법 스트리밍 플랫폼에서 유통 중인 영상 콘텐츠에 대해 왜곡 저항력이 강한 저작권 침해 여부 판단 기술을 제안한다. 제안하는 기법은 빠른 연산을 위해 수집된 데이터를 이진 데이터로 변환하여 원본 영상과의 해밍거리를 계산하는 방법을 사용하였으며, 실험 결과 평균 215KB의 크기와 94.79%의 정확도로 불법 복제 영상물을 판별할 수 있음을 확인하였다.

• 대한음성학회지:말소리
• /
• 제47호
• /
• pp.121-129
• /
• 2003

This paper describes the use of pitch information for speaker identification. The recognition system is a GMM based one with 4 connected Korean digits speech database. The mean of the pitch period in voiced sections of speech are shown to be ,useful at discriminating between speakers. Utilizing this feature with Gaussian mixture model in the speaker identification system gave a marked improvement, maximum 6% improvement comparing to the baseline Gaussian mixture model.

• 한국철도학회:학술대회논문집
• /
• 한국철도학회 2004년도 추계학술대회 논문집
• /
• pp.1390-1395
• /
• 2004

Recently, the load increasement and new regenerative systems of electrified railway system make it a difficult to distinguish between the load current and fault current. The failure of traction system perhaps causes over-current to flow. The high current can collapse other railway systems. If failures of the traction system takes place, the failures are detected and protected lest it should provoke high current flow. The over current from such a traction system failure permit to charge high tension voltage and produces high temperature arc, voltage instability, current cutting, and break down railway systems. The traction system failures detect and the system has to immediately cut off from over-current flow. To isolate the failure, the system can distinguish failure current from current flows. It forces us to adapt such as a new intelligent protection system. The protective system in traction system play a role of detecting and isolating failure points. In this paper, we proposed intelligent algorithm for discriminating normal and abnormal situation instead of the system being operated abnormally.