• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1467-1472
• /
• 2018

For the problem that the face image of surveillance video cannot be accurately identified due to the low resolution, this paper proposes a low resolution face recognition solution based on convolutional neural network model. Convolutional Neural Networks (CNN) model for multi-scale input The CNN model for multi-scale input is an improvement over the existing "two-step method" in which low-resolution images are up-sampled using a simple bi-cubic interpolation method. Then, the up sampled image and the high-resolution image are mixed as a model training sample. The CNN model learns the common feature space of the high- and low-resolution images, and then measures the feature similarity through the cosine distance. Finally, the recognition result is given. The experiments on the CMU PIE and Extended Yale B datasets show that the accuracy of the model is better than other comparison methods. Compared with the CMDA_BGE algorithm with the highest recognition rate, the accuracy rate is 2.5%~9.9%.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2009년도 추계학술대회
• /
• pp.738-741
• /
• 2009

본 연구는 잡음을 포함한 음성 환경에서의 음성인식을 개선방안에 관한 것이다. 우리는 음성인식에서 잡음 섞인 음성으로부터 얻은 스펙트럴 envelope에서 곡들의 스펙트럴 subtraction 및 복원이 보다 더 효과적임을 알 수 있었다. 본 연구에서, 평균화된 스펙트럴 envelope은 모음 스펙트럼으로부터 추출하여 곡들의 강조에 사용하였다. 낮은 주파수 영역에서의 모음 스펙트럴 정보는 강조되어지고 자음으로부터 얻은 스펙트럼은 변하지 않는다. 시뮬레이션으로 살펴보면, 강조계수는 켑스트럴 영역에서 변한다. 이 방법으로 잡음석인 숫자음성 인식에서 적용하였으며 인식결과가 개선됨을 알 수 있었다.

• 한국치위생학회지
• /
• 제15권4호
• /
• pp.593-602
• /
• 2015

Objectives: The purpose of the study is to investigate the relations between oral health status and subjective oral health recognition in Korean adolescents. Methods: The survey data were extracted from the 2012 Korea National Oral Health Survey from June to November, 2012. The survey consisted of oral checkup data and the oral health interviewing data of 9,981 adolescents(5,335 male, 4,646 female) by proportional distribution. The questionnaire included general objective oral health status and subjective oral health recognition. The subjective oral health recognition consisted of general characteristics of the subjects, oral health status, oral health behavior, and subjective oral health recognition. Results: According to multiple regression analysis for the relations between general characteristics and subjective oral health recognition, subjective health recognition, the influencing factors were gender, age, the number of tooth brushing per day, intake of snacks and carbonated drinks, regular oral examination, decayed teeth (DT), missing teeth (MT,) and DMFT. Conclusions: The routine oral checkup is the most important method to improve the oral health management in the adolescents. The continuing oral health improvement project must be implemented based on the law and will keep the adolescents in good oral health status.

• 디지털산업정보학회논문지
• /
• 제8권1호
• /
• pp.29-40
• /
• 2012

A challenging research issue that has been one of growing importance to those working in human-computer interaction are to endow a machine with an emotional intelligence. Thus, emotion recognition technology plays an important role in the research area of human-computer interaction, and it allows a more natural and more human-like communication between human and computer. In this paper, we propose the multimodal emotion recognition system using face and speech to improve recognition performance. The distance measurement of the face-based emotion recognition is calculated by 2D-PCA of MCS-LBP image and nearest neighbor classifier, and also the likelihood measurement is obtained by Gaussian mixture model algorithm based on pitch and mel-frequency cepstral coefficient features in speech-based emotion recognition. The individual matching scores obtained from face and speech are combined using a weighted-summation operation, and the fused-score is utilized to classify the human emotion. Through experimental results, the proposed method exhibits improved recognition accuracy of about 11.25% to 19.75% when compared to the most uni-modal approach. From these results, we confirmed that the proposed approach achieved a significant performance improvement and the proposed method was very effective.

• 디지털융복합연구
• /
• 제10권11호
• /
• pp.359-363
• /
• 2012

어휘 인식 모델에서는 정확하지 않은 어휘로 부터 특징을 추출하기 때문에 어휘가 실제 어휘와 유사한 어휘로 인식되거나 인식이 되지 않는 현상이 나타난다. 이를 위해 본 논문에서는 효율적인 형상 형성을 지원하는 시스템을 모델링하고 구현하였으며, 형상 형성 정보를 효율적으로 처리하고 어휘 유사율 관리를 최적화하기 위해 데이터베이스 검색에서 facet 방법을 응용하였다. 본 논문에서 제안한 시스템을 적용한 결과 시스템 성능에서 어휘 종속 인식률은 95.31%, 어휘 독립 인식률은 97.38%의 인식률을 나타내었다.

• 공학교육연구
• /
• 제20권5호
• /
• pp.74-86
• /
• 2017

The purpose of this study is to analyze the perceptions of college officials about the evaluation of university education on the perspective of industry, to make policy proposals for improvement by analyzing the preparing process of the departments that have received the evaluation of the best universities, In order to achieve the purpose of the study, we conducted literature review, survey studies, and in-depth interviews. The survey was conducted by professors related to this system, and finally the results of 51 responses were used for analysis. The contents of the survey were composed of the recognition, appropriateness, the utilization of the results, the proliferation and improvement plan of the evaluation of university education on the perspective of industry. In-depth interviews conducted 1:1 interviews with 5 professors. The evaluation of university education on the perspective of industry is generally perceived to be a bridge between industry and university education, but there is room for improvement in evaluation criteria, operating methods, and promotion strategies to public.

• 보건의료산업학회지
• /
• 제3권2호
• /
• pp.29-43
• /
• 2009

The purpose of this study is to provide basic material about the improvement method of the hospital evaluation system by grasping recognition, corresponding behavior to evaluation, influence and problems about the hospital evaluation system with respect to the employee of mid sized hospitals having relatively poorer environment. In this study were used 518 questionnaires to hospital employee working at 20 general hospitals in Pusan, Kyungnam with less than 500 beds and having experience of the hospital evaluation during 2005~2007. As a result of the research, they felt the investigation tool used for the hospital evaluation has too many items and it could be difficult to reflect characteristics of each hospital. It appeared that utilization of the report result after the hospital evaluation was helpful for grasping a part necessary for hospital improvement and the present level of hospitals and increase of interest about medical service quality improvement of the a hospital director or hospital employee. They trusted the result of the hospital evaluation to some degree, and recognized it helps the patients to select a hospital and increase rights-to-know. Hospital employee recognized the necessity and showed high interest and participation, but they didn't sympathize in the propriety and the feasibility of the hospital evaluation items. High degree evaluated hospital employee receive hospital evaluation system positively and think hospital services to be improve after taking hospital evaluation. The hospital evaluation should make the hospital to effort to achieve medical quality improvement to its purpose. Also, systematic reformation will be necessary to reflect characteristics of the hospital that has relatively small scale or locates in poor environment.

• 한국환경교육학회지:환경교육
• /
• 제22권1호
• /
• pp.97-109
• /
• 2009

The purpose of this study is to grasp the degree of recognition of the ESD(Education for Sustainable Development) form Technology Home Economic teachers in high schools, to analyze contents related to the ESD within the textbooks, and to propose the improvement method of the textbooks. As a result, first, the degree of recognition on the ESD was less than about 30.1% in Technology Home Economic teachers. Second, the content and factor criterions for the ESD were developed on the basis of the previous studies. They are suitable for applying to the Vehicle Chapter in the textbooks and are including 9 contents and 7 factors, respectively. Third, the contents related to the ESD within a Technology. Home Economic textbook showed 36.1% in Technology, 33.8% in Home Economics including social and cultural(44.7%), environmental(47.1%), and economic contents(8.2%). But they in the Vehicle Chapter within the textbooks presented 69.1%, 4.8%, 26.1% in social and cultural, environmental, and economic contents, respectively. Finally, the improvement method of the new textbook for the Vehicle Chapter is to adopt issued factors including the suitable adjustment on the vehicle structure and safety as well as the supplement of environmental contents related to the ESD. In addition, it is necessary to contain the contents such as the developing process of environment-friendly vehicles, the economically sustainable consumption strategy, and the cause-and-effect relationship with the global warming in the new textbook.

• 한국전자통신학회논문지
• /
• 제8권1호
• /
• pp.77-83
• /
• 2013

본 논문에서는 음성 인식률 개선에 관한 방법을 제시하고 연구하였다. 기존의 음성 검출 방법 중 많이 이용되고 있는 HMM(Hidden Markov Model) 알고리즘을 이용하여서 음성을 검출하였다. 실험은 음성 검출과 음성 인식의 두 가지 방법으로 진행하였다. 음성 검출은 음성의 단위로 영교차율을 구하여 데이터의 유무를 판별하였다. 음성 인식은 음성의 형상의 패턴을 분석한 후 학습된 패턴과 비교 하는 형식으로 분석하였다. 실험 결과, 제안된 음성 형상의 패턴인식 이용한 알고리즘은 92%의 음성 인식률을 얻어 80%의 기존 HMM 알고리즘에 비해서 약 12%의 향상된 인식률을 얻을 수 있었다.

• 한국통신학회논문지
• /
• 제28권9C호
• /
• pp.876-883
• /
• 2003

본 논문에서는 초음파 영상에서 로그 전력 스펙트럼(log power spectrum)을 이용한 장기 인식 알고리듬을 제시한다. 제안한 알고리듬은 크게 특징추출과 특징분류의 두 단계로 구성된다. 특징추출에서는 이동불변의 성질을 가지는 로그 전력 스펙트럼을 이용하여 전처리를 수행한 입력 영상으로부터 장기 조직의 반향(echo of the tissue) 성분을 추출한다. 특징 분류에서는 마하라노비스(Mahalanobis) 거리를 사용하여 입력영상으로부터 추출한 특징벡터와 각 영상 부류의 평균벡터 사이의 유사도를 측정한다. 실제 초음파 영상에 대한 실험결과는 제안된 알고리듬이 전력 스펙트럼(power spectrum)과 유클리드(Euclid) 거리를 이용한 인식 알고리듬보다 최대 30% 향상된 인식률을, 또 가중 큐프런시(weighted quefrency) 복소 켑스트럼(complex cepstrum)을 이용한 알고리듬보다 10∼40% 향상된 인식률을 보여준다.