• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2003.06a
• /
• pp.46-49
• /
• 2003

The use of confidence measures for word/utterance verification has become art essential component of any speech input application. Confidence measures have applications to a number of problems such as rejection of incorrect hypotheses, speaker adaptation, or adaptive modification of the hypothesis score during search in continuous speech recognition. In this paper, we present a new utterance verification method using vowel string. Using subword HMMs of VCCV unit, we create anti-models which include vowel string in hypothesis words. The experiment results show that the utterance verification rate of the proposed method is about 79.5%.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.229-230
• /
• 2020

컴퓨터 하드웨어, 소프트웨어의 급속한 발전에도 불구하고 인간의 정보처리능력에 뒤지는 경우도 있다. 이와 같이 인간의 정보처리체계가 어떻게 이런 문제를 해결할 수 있는지에 대한 연구가 활발히 진행 중이다. 본 논문에서는 인간의 시각 정보 처리의 특성에 의한 인지 정보 처리 모델링을 기반으로 학습 가능성을 보이고 일반적인 영상 인식 모델과의 차이를 비교 분석하고자 한다.

• Journal of Embryo Transfer
• /
• v.31 no.2
• /
• pp.137-143
• /
• 2016

This study was conducted to evaluate a visual surveillance system. The advancement of recording technology and network service make it easy to record and transfer the videos. Moreover, progressed recognition technology help to make a distinction each other. Cows show distinguishing behaviors during their estrus period. The mounting is one of the behaviors. The result was different depending on the breed of the cows and the size of the farm. In the case of Korean native cattle, the estrus detection rate was 71.15%, however, dairy cows, the estrus detection rate was 39.38%. At the farms having below 6 modules, the estrus detection rate was 87.41%. On the other hand, at the farms having over 6 modules, the estrus detection rate was 77.78%. With the proper progress, the visual surveillance system can be used to detect heat detection.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.11
• /
• pp.1408-1414
• /
• 2018

Recently, as a solution to the problem of maintaining system reliability, stability, and quality occurring worldwide, such as activation of smart grid and recognition of super grid and rapid grid interconnection of renewable energy sources HVDC(High Voltage Direct Current) will appear on the front of the electric power system. These concepts are also very important concepts in HVDC systems. When the HVDC system is linked to the existing power system, it is composed of AC/DC/AC conversion device, and these conversion devices are composed of many thyristors. These parts(Devices) are connected in a complicated manner, and they belong to the one with a higher failure rate. However, the problem of establishing the concept of failure rate of HVDC parts directly linked to economic efficiency and the understanding accompanying it are still insufficient. Therefore, in this paper, we establish the meaning of reliability in power system and try to develop a model to analyze and verify the failure rate data of HVDC based on this.

• Health Policy and Management
• /
• v.26 no.4
• /
• pp.382-389
• /
• 2016

Background: National cancer screening reduces mortality from stomach, colon, breast, and cervical cancer. However, cancer screening rates remain lower than 50% in Korea. This research was designed to make effective guidelines for cancer screening promotion to increase participation rate in cancer screening. Methods: Data was collected from Incheon regional cancer center's telephone survey 'Recognition of cancer prevention and control' of 861 Incheon citizens over age 40. Logistic regression was used to analyse a relationships between cancer screening rates and promotional methods. PASW SPSS ver. 18.0 program (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. Results: About 27.4% of respondents (236 of 861 respondents) have recognized the promotional program of Incheon regional cancer center for cancer screening participation. Cancer screening rate was associated with age group, educational level for stomach cancer, age group, bus advertisement (odds ratio, 1.19 to 2.75) for colon cancer, educational level, residential area for breast cancer, and age group for cervical cancer. Conclusion: There was relationship between stomach cancer screening rates and promotional methods. Therefore, cancer screening promotion guidelines to increase participation rate should be considered.

• Child Health Nursing Research
• /
• v.22 no.3
• /
• pp.172-181
• /
• 2016

Purpose: This study was conducted to examine the effect of education on infection control for multidrug resistant organism (MDRO). Methods: One group pre-post time series design was used. Infection control education for MDRO infection was provided to nurses working in the Neonatal Intensive Care Unit (NICU). Knowledge and recognition were evaluated before and after education. Hand hygiene compliance, MDRO isolation rate and central line-associated bloodstream infection (CLABSI) rate were used as outcome variables. Data from 45 nurses was used for analysis. Results: General knowledge about MDRO increased (p=.011). Responses to questions about image of MDRO and MDRO management tended to change in a positive direction (p=.046). Hand hygiene compliance was 100% at pre-test, 79.5% during education period and 98.4% at post-education period. MDRO isolation rate was 6.83 per 1,000 patient days at pre-test, 10.24 during education period and 6.68 at post-education period. CLABSI rate was 3.76 per 1,000 central line days at pre-test, 6.84 during education period and 4.71 at post-education period. Conclusion: Findings indicate that the education program is effective in improving knowledge about MDRO in NICU nurses. However, more reliable indicators should be used to determine long-term effects.

• Journal of Korea Multimedia Society
• /
• v.18 no.1
• /
• pp.88-96
• /
• 2015

The recent increase in the production of 3D contents allowed viewers to experience various 3D contents. However, some of the viewers did not experience 3D depth well. Several researches were done in past to measure viewers' 3D depth perception, but these researches were done with certain limitations. In this paper, we measured viewers' 3D depth perception and recovery rate in relation with the changes in binocular disparities, saturation, and brightness values after subjects' watching 2D/3D contents. The results showed that when viewers watched the 3D content with positive binocular disparities for 42 minutes, viewers felt that the object seemed to have moved further forward than it was before; with 3D content with negative binocular disparities, viewers felt that the object seemed to be moved backwards. We found that the locational differences of the object in positive disparities were greater than those in the negative binocular disparities. The recovery rate was computed by comparing two measured values of before and after watching 3D contents for 30 minutes. On average, after 30-minute break, viewers showed roughly 50 % of recovery rate.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.29 no.4
• /
• pp.28-36
• /
• 2021

Surprise and Startle are often cited as potential or contributing factors in the cause of aircraft accidents because they can negatively affect pilots' situational awareness and decision-making. Accordingly, it is necessary to more clearly understand the effect on flight safety through the psychological and physiological reactions that occur to the pilots. Therefore, this study empirically analyzed the effects of Surprise and Startle on the pilot's psychological and physiological responses using a helicopter flight training device for the first time in Korea. As a result of the analysis, the ratio of recognition of Startle and Surprise, heart rate, and respiration rate showed a high difference between private pilots and commercial pilots. This is analyzed as the difference between long-term flight experience and flight time. In addition to presenting effective information for improving the education and training field of helicopter pilots in the future, it is intended to provide basic data for related research.

• Journal of Korean society of Dental Hygiene
• /
• v.23 no.4
• /
• pp.303-309
• /
• 2023

Objectives: According to the Korea Education Development Institute, the college admission quota is expected to exceed the number of high school graduates, leading to an anticipated expansion in the gap between them. This paper aims to conduct an empirical analysis of the variables previously studied, with a specific focus on the admission competition rate for the department of dental hygiene. Methods: The research methodology is the multiple linear regression analysis. The research data contains the structured data from academy information, and the web-based unstructured data collected over the past 3 years. Results: After conducting the analysis, it was newly discovered that the university's online recognition and its location in the metropolitan area were statistically significant factors influencing the admission competition rate for the department of dental hygiene. Conclusions: The findings of this study are expected to be helpful in formulating admission strategies for universities to attract new students and identifying the factors that influence student attraction.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.453-462
• /
• 2021

A lot of license plate data is required for car number recognition. License plate data needs to be balanced from past license plates to the latest license plates. However, it is difficult to obtain data from the actual past license plate to the latest ones. In order to solve this problem, a license plate recognition study through deep learning is being conducted by creating a synthetic license plates. Since the synthetic data have differences from real data, and various data augmentation techniques are used to solve these problems. Existing data augmentation simply used methods such as brightness, rotation, affine transformation, blur, and noise. In this paper, we apply a style transformation method that transforms synthetic data into real-world data styles with data augmentation methods. In addition, real license plate data are noisy when it is captured from a distance and under the dark environment. If we simply recognize characters with input data, chances of misrecognition are high. To improve character recognition, in this paper, we applied the DeblurGANv2 method as a quality improvement method for character recognition, increasing the accuracy of license plate recognition. The method of deep learning for license plate detection and license plate number recognition used YOLO-V5. To determine the performance of the synthetic license plate data, we construct a test set by collecting our own secured license plates. License plate detection without style conversion recorded 0.614 mAP. As a result of applying the style transformation, we confirm that the license plate detection performance was improved by recording 0.679mAP. In addition, the successul detection rate without image enhancement was 0.872, and the detection rate was 0.915 after image enhancement, confirming that the performance improved.