• Title/Summary/Keyword: recognition-rate

Search Result 2,809, Processing Time 0.029 seconds

Study for an Artificial Visual Machine for the Blind (맹인용인공시각보조장치에 관한 연구)

  • 홍승홍;이균하
    • Journal of the Korean Institute of Telematics and Electronics
    • /
    • v.15 no.5
    • /
    • pp.19-24
    • /
    • 1978
  • In this paper, the functional propertied of vibrotactile sense of skin were studied by means of psycophysical experiments with respect to frequency and waveform of mechanical vibration, two-point threshold, contactor size of stimulators. Furthermore, leased on the experimental result, a small vibrotactile stimulator made of piezoelectrc ireed vibrator array was proposed for a aid blind to recognition of the Korean letters. A tactile output image is presented by an 8 row$\times$1 column array of samall vibrator reeds with 200 Hz rectangular wave, the array fitting on a fore-finger. Under the control of the NOVA mini-computer, the bimorph reeds array could represent any of one of the 24 characters of the Korean vowel and consonant at the 8 positions from left to right on the array. Without learning effect, the identification test of the Korean characters by the designed experimental system was carried out. The average rate of correct response was 90%.

  • PDF

Design and Implementation of a Subjective-type Evaluation System Using Syntactic and Case-Role Information (구문-격의미 정보를 이용한 주관식 문제 채점 시스템 설계 및 구현)

  • Kang, Won-Seog
    • The Journal of Korean Association of Computer Education
    • /
    • v.10 no.5
    • /
    • pp.61-69
    • /
    • 2007
  • The subjective-type evaluation can estimate the high-recognition ability, but has the problem of the objectivity and reliability of the evaluation, and the difficulty of Korean language processing. To solve the problem, this paper designs and implements a subjective-type evaluation system using syntactic and case-role information. This system can reduce the time and endeavor for evaluation and provide the objectivity of the evaluation. The system results the 75% success rate to the instructor evaluation and gets the better precision and recall than the word extraction evaluation system. We expect that this system will become a basis of the research on the subjective-type evaluation.

  • PDF

Psychosocial Work Environment and Self-rated Health of Nurses in a General Hospital (일개 종합병원 간호사의 사회심리적 업무환경과 주관적 건강)

  • Choi, Eunsuk;Lee, Yangsun
    • Korean Journal of Occupational Health Nursing
    • /
    • v.23 no.4
    • /
    • pp.219-226
    • /
    • 2014
  • Purpose: This study aims to determine the association between psychosocial work environment and self-rated health among general hospital nurses. Methods: A total of 195 nurses working in one general hospital were eligible for data analysis by multivariate logistic regression. The psychosocial work environment was measured with the Korean version of the Copenhagen Psycosocial Questionnaire version II (COPSOQ-K). Self-rated health was recoded as good (excellent/good) and not good (fair/poor/bad) to the question, "In general, how would you rate your health status?" Results: 40% of nurses rated their health positively. Commitment to the workplace (OR=1.27), predictability (OR=1.32), recognition and reward (OR=1.41), role clarity (OR=1.32), and social support from colleagues (OR=1.25) were positively associated with self-rated health of nurse participants. Work-family conflict (OR=0.82) was negatively associated with self-rated health. Conclusion: The findings suggest that psychological work environment predicts self-rated health of hospital nurses. Good psychological work environment may be helpful in improvement of nurses' health.

Development of Parameters for Diagnosing Laryngeal Diseases

  • Kim, Yong-Ju;Wang, Soo-Geun;Kim, Gi-Ryun;Kwon, Soon-Bok;Jeon, Kye-Rok;Back, Moo-Jin;Yang, Byung-Gon;Jo, Cheol-Woo;Kim, Hyung-Soon
    • Speech Sciences
    • /
    • v.10 no.1
    • /
    • pp.117-129
    • /
    • 2003
  • Many people suffer from various laryngeal diseases. Since we can notice voice change easily, acoustic analysis can be helpful to diagnose the diseases. Several attempts have been made to clarify the relation between the parameters and the state of sick vocal folds but any decisive parameters are not found yet. The purpose of this study was to select and develop those parameters useful for diagnosing and differentiating laryngeal diseases. We examined eight MDVP parameters, and two additional MFCC and LPC parameters obtained from the production of an open vowel by 252 subjects with or without laryngeal diseases. Using a statistical procedure through the artificial neural networks, we attempted to differentiate laryngeal disease groups. Results showed that the LPC parameters indicated the highest differentiating rate by the networks followed by the MFCC and the MDVP parameters. In addition, Jita, Shim and NHR among the MDVP parameters came out better parameters in diagnosing laryngeal diseases.

  • PDF

L1-norm Regularization for State Vector Adaptation of Subspace Gaussian Mixture Model (L1-norm regularization을 통한 SGMM의 state vector 적응)

  • Goo, Jahyun;Kim, Younggwan;Kim, Hoirin
    • Phonetics and Speech Sciences
    • /
    • v.7 no.3
    • /
    • pp.131-138
    • /
    • 2015
  • In this paper, we propose L1-norm regularization for state vector adaptation of subspace Gaussian mixture model (SGMM). When you design a speaker adaptation system with GMM-HMM acoustic model, MAP is the most typical technique to be considered. However, in MAP adaptation procedure, large number of parameters should be updated simultaneously. We can adopt sparse adaptation such as L1-norm regularization or sparse MAP to cope with that, but the performance of sparse adaptation is not good as MAP adaptation. However, SGMM does not suffer a lot from sparse adaptation as GMM-HMM because each Gaussian mean vector in SGMM is defined as a weighted sum of basis vectors, which is much robust to the fluctuation of parameters. Since there are only a few adaptation techniques appropriate for SGMM, our proposed method could be powerful especially when the number of adaptation data is limited. Experimental results show that error reduction rate of the proposed method is better than the result of MAP adaptation of SGMM, even with small adaptation data.

A Study on the eco-friendly design curriculum in the interior design education - Focus on the curriculum of 4year Universities - (실내디자인 교육의 친환경 교육과정 현황에 관한 연구 -4년제 대학교의 교육과정을 중심으로-)

  • Lee, Kyung Eun;Huh, Sung Eun
    • KIEAE Journal
    • /
    • v.7 no.6
    • /
    • pp.45-52
    • /
    • 2007
  • The concern and recognition of the energy, environment, conservation of natural resources of the world today are raising, and sustainable development can't be considered in a way of separating development from environment any more. For sustainable development, a new educational infrastructure construction and paradigm coping with new environment through the whole education are needed. Accordingly, this research analyzed eco-friendly design curriculum among interior design major of 4year universities to help with reestablishing educational goal as a model for eco-friendly design curriculum development. According to this research, environment related subject were 91, the rate of 16.4% among total 1492 interior designs major, and among them, eco-friendly design related subjects were 15, which is the 6.4% of 91 environment related subjects, and which is 1.0% among total interior designs, through which we could see that eco-friendly design related subjects were very light weight in the total interior designs. So, the reasonable and concrete reform of eco-friendly design related subjects in 4year universities interior design curriculum seems to be urgent, and diverse eco-friendly design related subjects needed to be developed and established.

IR LED Marker Detection Method for Production of Multiple Marker based on Augmented Reality (다수 마커의 제작을 위한 증강현실 기반의 IR LED 마커 검출 기법)

  • Lee, Hye-Mi;Ryu, Nam-Hoon;Kim, Eung-Kon
    • The Journal of the Korea institute of electronic communication sciences
    • /
    • v.6 no.3
    • /
    • pp.457-463
    • /
    • 2011
  • As computer related technologies are developed, interests in augmented reality technologies are greatly increasing. Augmented reality is a technology that composes digital contents from the real input images through camera and it enables interaction with users. This study designed a directional marker using LED light that emits infrared ray, then, provided a detection algorithm and a marker information extraction method that can realize various virtual objects as augmented reality from one marker. The newly designed method provides a solution to settle the problems in existing marker technologies such as decrease of immersiveness and read rate and single information expression, and at the same time it can minimize the cost or time consumption in marker information storage.

Design of Sajik Fountain (사직분수대 설계)

  • 김영인;김민중;김정수;홍형순;장태현
    • Journal of the Korean Institute of Landscape Architecture
    • /
    • v.29 no.4
    • /
    • pp.75-81
    • /
    • 2001
  • This project provides designs for renewal of the surrounding environment related to old fountains typically located in cities. In many cities, the fountain element has bee located with a strategic position in the city. Located in the central area of Cheongju, the Sajik Fountain Fountain has significant position in the local city. However, due to changes in the surrounding area and the old-fashioned form of the fountain, it has not contributed substantially to the cityscape. Including the historical Seomoon Bridge and Moosim-Cheon(stream) as Wooam-San(mountain) setting, the surrounding environment of our planning site appear to be a locally important place. In respect to building with a broad function, landscape designers focused on the reconsideration of symbolization as landmark, of the image of Cheongju city, of recognition by removing the horizontal line and of the substantial civil´s use rate. Achieved by working with a sculptor, a new function, the main facility, is managed sculpturally and imaged in a drop of water representing clear, fresh Cheongju city. This plan introduces the image of Wooam-San, Moosim-Cheon(stream), Sangdang-SanSung(A mountain fortress) and elicit a sense of peace, symbolic of Cheongju. The significance of this project is that as the current cities are increasingly developed, small areas left alone in city are reconsidered and offered to the public as efficient, resting and active outdoor space.

  • PDF

Analysis of Human Activity Using Motion Vector (움직임 벡터를 이용한 사람 활동성 분석)

  • Kim, Sun-Woo;Choi, Yeon-Sung;Yang, Hae-Kwon
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 2011.10a
    • /
    • pp.157-160
    • /
    • 2011
  • In this paper, We proposed the method of recognition and analysis of human activites using Motion vector in real-time surveillance system. We employs subtraction image techniques to detect blob(human) in the foreground. When MPEG-4 video recording EPZS(Enhanced Predicted Zonal Search) is detected the values of motion vectors were used. In this paper, the activities of human recognize and classified such as meta-classes like this {Active, Inactive}, {Moving, Non-moving}, {Walking, Running}. Each step was separated using a step-by-step threshold values. We created approximately 150 conditions for the simulation. As a result, We showed a high success rate about 86~98% to distinguish each steps in simulation image.

  • PDF

Head/Rear Lamp Detection for Stop and Wrong Way Vehicle in the Tunnel (터널 내 정차 및 역주행 차량 인식을 위한 전조등과 후미등 검출 알고리즘)

  • Kim, Gyu-Yeong;Do, Jin-Kyu;Park, Jang-Sik;Kim, Hyun-Tae;Yu, Yun-Sik
    • Proceedings of the Korean Institute of Information and Commucation Sciences Conference
    • /
    • 2011.10a
    • /
    • pp.601-602
    • /
    • 2011
  • In this paper, we propose head/rear lamp detection algorithm for stopped and wrong way vehicle recognition. It is shown that our algorithm detected vehicles based on the experimental analysis about the color information of vehicle's lamps. The simulation results show the detection rate about stopped and wrong way vehicles is achieved over 94% and 96% in the tunnel HD(High Definition) video image.

  • PDF