• Proceedings of the KSPS conference
• /
• 2004.05a
• /
• pp.239-242
• /
• 2004

An automatic pronunciation correction system provides users with correction guidelines for each pronunciation error. For this purpose, we propose a speech recognition system which automatically classifies pronunciation errors when Koreans speak a foreign language. In this paper, we also propose machine scoring methods for automatic assessment of pronunciation quality by the speech recognizer. Scores obtained from an expert human listener are used as the reference to evaluate the different machine scores and to provide targets when training some of algorithms. We use a log-likelihood score and a normalized log-likelihood score as machine scoring methods. Experimental results show that the normalized log-likelihood score had higher correlation with human scores than that obtained using the log-likelihood score.

• Speech Sciences
• /
• v.11 no.2
• /
• pp.217-226
• /
• 2004

An automatic pronunciation correction system provides users with correction guidelines for each pronunciation error. For this purpose, we develop a speech recognition system which automatically classifies pronunciation errors when Koreans speak a foreign language. In this paper, we propose a machine scoring method for automatic assessment of pronunciation quality by the speech recognizer. Scores obtained from an expert human listener are used as the reference to evaluate the different machine scores and to provide targets when training some of algorithms. We use a log-likelihood score and a normalized log-likelihood score as machine scoring methods. Experimental results show that the normalized log-likelihood score had higher correlation with human scores than that obtained using the log-likelihood score.

• Smart Media Journal
• /
• v.8 no.4
• /
• pp.64-71
• /
• 2019

The performance of an OMR (Optical Music Recognition) system is usually determined by the characterizing features of the input music score images. Low resolution is one of the main factors leading to degraded image quality. In this paper, we handle the low-resolution problem using the super-resolution technique. We propose the use of a deep neural network with instance normalization to improve the quality of music score images. We apply instance normalization which has proven to be beneficial in single image enhancement. It works better than batch normalization, which shows the effectiveness of shifting the mean and variance of deep features at the instance level. The proposed method provides an end-to-end mapping technique between the high and low-resolution images respectively. New images are then created, in which the resolution is four times higher than the resolution of the original images. Our model has been evaluated with the dataset "DeepScores" and shows that it outperforms other existing methods.

• Korean Journal of Biological Psychiatry
• /
• v.21 no.1
• /
• pp.21-27
• /
• 2014

Objectives The aim of the present study was to investigate gender difference in empathic ability and recognition of facial emotion expression in schizophrenic patients. Methods Twenty-two schizophrenic outpatients (11 men and 11 women) and controls (10 men and 12 women) performed both the scale of Empathic Quotient (EQ) and facial emotion recognition test. We compared the scores of EQ and the facial emotion recognition test among each group according to diagnosis and gender. Results We found a significant sex difference in the scores of EQ and the facial emotion recognition test in the schizophrenic patients. And there were significantly negative correlation between the score of the facial emotion recognition test and the scores of Positive and Negative Symptom Scale (PANSS) in female schizophrenic patients. However, in male schizophrenic patients, there were no significant correlations between the score of each test and the scores of PANSS. Conclusions This study suggests that the sex difference in empathic ability and facial emotion recognition would be very important in chronic schizophrenic patients. Investigation of sex effects in empathic ability and facial emotion recognition in chronic schizophrenic patients would present an important solution for constructing optimal rehabilitation program.

• The Korean Journal of Health Service Management
• /
• v.6 no.2
• /
• pp.121-131
• /
• 2012

This study investigated the difference of the university students' recognition, attitude, and knowledge level of health by the health education in Daejeon and Chungnam province. First, in the survey analysis result regarding the health related recognition, the students who had attended lectures of health education were surveyed not only to have higher health related recognition than the students who did not have attended the lecture of health education, but also to appear meaningful in statistics by showing significance level p<0.05 in total score. Second, in the research regarding the health related attitude, the attendance or nonattendance of health education lecture were surveyed not to have high effect on the attitude of daily lives of the university students. Third, in the analysis of health related knowledge, the knowledge level between the students who had attended lectures of health education and who did not have attended the lecture of health education showed that the students who had attended lectures of health education got more excellent correct answer rate. In this survey result, the students who had attended lectures of health education showed higher health related recognition and knowledge than the students who had not attended lectures of health education.

• Journal of the Korean Society of Industry Convergence
• /
• v.25 no.6_3
• /
• pp.1285-1291
• /
• 2022

In this paper, we propose a low-cost, low-power embedded environment-based deep learning lightweight model for input images to recognize laundry management codes. Laundry franchise companies mainly use barcode recognition-based systems to record laundry consignee information and laundry information for laundry collection management. Conventional laundry collection management systems using barcodes require barcode printing costs, and due to barcode damage and contamination, it is necessary to improve the cost of reprinting the barcode book in its entirety of 1 billion won annually. It is also difficult to do. Recognition performance is improved by applying the VGG model with 7 layers, which is a reduced-transformation of the VGGNet model for number recognition. As a result of the numerical recognition experiment of service parts drawings, the proposed method obtained a significantly improved result over the conventional method with an F1-Score of 0.95.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.321-323
• /
• 2016

최근 연구에서 기계학습 중 지도학습 방법으로 개체명 인식을 하고 있다. 그러나 지도 학습 방법은 데이터를 만드는 비용과 시간이 많이 필요로 한다. 본 연구에서는 주석 된 말뭉치를 사용하여 지도 학습 방법을 사용 한다. 의생명 개체명 인식은 Protein, RNA, DNA, Cell type, Cell line 등을 포함한 텍스트 처리에 중요한 기초 작업입니다. 그리고 의생명 지식 검색에서 가장 기본과 핵심 작업 중 하나이다. 본 연구에서는 순환형 신경망과 워드 임베딩을 자질로 사용한 조건적 임의 필드에 대한 성능을 비교한다. 조건적 임의 필드에 N_Gram만을 자질로 사용한 것을 기준점으로 설정 하였고, 기준점의 결과는 70.09% F1 Score이다. RNN의 jordan type은 60.75% F1 Score, elman type은 58.80% F1 Score의 성능을 보여준다. 조건적 임의 필드에 CCA, GLOVE, WORD2VEC을 사용 한 결과는 각각 72.73% F1 Score, 72.74% F1 Score, 72.82% F1 Score의 성능을 얻을 수 있다.

• Journal of the Korean Society of School Health
• /
• v.16 no.2
• /
• pp.111-123
• /
• 2003

The purpose of this study was to identify the expectation about school health program of primary school parents as an important variable for growth of school health program fore area locked primary school parents 631 by self-administered questionnaire between June 20 and September 29, 2001 The results were as follows 1. General characteristic of the Subjects Age rates of the subjects were 45.5% on 35~39 years old and 34.7% over 40 years old. Economic status were middle level 86%, participation rates of school health education for parent were 16.8%, recognition rates on significance of school health education were 89.9%. 2. Health knowledge of parents The mean score on health knowledge of the subjects was 6.8 from a total of 10. there was a statistically significant difference with regard to health knowledge among the 7school(F=3.46, p=.016). 3. Health behavior of parents The mean score on health behaviors of the subjects was 77.3 from a total of 100. there was no statistically difference in health behavior among the school. 4. The expectation about school health program of the parents The mean score on the expectation about school health program of the subjects was 51.1 from a total of 140. there was a statistically significant difference with regard to the expe ctation about school health program among the school(F=3.75, p=.010). 5. Health behaviors and the expectation about school health program from sociod emogr aphic Backgrounds of the Subjects. In the backgrounds of the subjects, the score on health behavior was higher the mothers(t=2.89,p=.004), the upper class(F=7.38, p=.000), the higher participation level of school health education of parent(F=7.38, p=.000) and the higher recognition level on significance of school health education of parent(F=9.94, p=.000). The score on the expectation about school health program was higher the lower recognition level on significance of school, health education of parent and on the expectation about health screening & preventive services was significant difference by age. however, the difference was not significant in the others.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.45 no.5
• /
• pp.192-200
• /
• 2008

Multimodal biometrics is a method for personal authentication and verification using more than two types of biometrics data. RBF based score-level fusion uses pattern recognition algorithm for multimodal biometrics, seeking the optimal decision boundary to classify score feature vectors each of which consists of matching scores obtained from several unimodal biometrics system for each sample. In this case, all matching scores are assumed to have the same reliability. However, in recent research it is reported that the quality of input sample affects the result of biometrics. Currently the matching scores having low reliability caused by low quality of samples are not currently considered for pattern recognition modelling in multimodal biometrics. To solve this problem, in this paper, we proposed the RBF based score-level fusion approach which employs quality information of input biometrics data to adjust decision boundary. As a result the proposed method with Qualify information showed better recognition performance than both the unimodal biometrics and the usual RBF based score-level fusion without using quality information.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.4
• /
• pp.1703-1710
• /
• 2011

Objectives: Using a PHR(Personal Health Record) is very important for the management of chronic disease and health. The young age group in this study was asked to pretend that they were members in the old age group (adult group here after) in order to investigate the recognition level differences in such conditions. Methods: We conducted face to face surveys with two age groups. 131 Adults and 398 University students, from May 11 to 22 in 2009. The questionnaires were composed of 18 items. Results & Conclusion: Adults replied more favorably than University students in using PHR willingness(University group 3.3score, Adults 3.7score) and recommendation(university 3.1score, Adults 3.8score). Adults liked paper PHR (63.2%), whereas, University students favored electronic Personal Health Record (71.1%). University students were highly concerned about disclosing the information(4.5score) of their PHR. They thought the appropriate time of education for PHR is high school or University degree. Therefore, in the future, it is vital that a PHR for young age group should be begun in early education and that an ePHR should be developed.