• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04b
• /
• pp.532-534
• /
• 2001

본 논문에서는 다양한 배경을 가지는 얼굴 영상에서 눈의 위치를 추출하고 누의 개폐 상태를 인식하는 방법에 대하여 제시한다. 얼굴 요소 중에서 눈은 얼굴 인식 분야에 있어서 주요한 특징을 나타내는 주 요소이며, 눈의 개폐 상태 인식은 인간의 물리적, 생체적 신호 감지 및 표정인식에도 유용하게 사용될 수 있다. 본 논문에서는 후부영역을 강조하기 위한 전처리 과정을 수행하고 템플릿 매칭 방법을 사용하여 후부 영역을 추출한다. 추출된 1차 후부 영역들은 설정된 병합식을 사용하여 병합되며, 기하학적 사전지식과 Matching Value를 기반으로 최종 눈후보 영역을 추출한다. 검출된 눈 후보 영역은 검출영역 전처리와 특징점 산출 과정을 거쳐 최종적으로 개폐 판별식을 통해 눈의 개폐상태를 인식하게 된다. 제안한 방법은 눈위치 추출과 개폐인식에서 모두 높은 인식률을 보였으며 향후 운전자의 졸음인식 및 환자 감시장치 등 여러 응용에서 사용될 수 있다.

• Proceedings of the KIEE Conference
• /
• 2002.07c
• /
• pp.1698-1700
• /
• 2002

Recently, the HFPD measurement methods that detect high frequency bandwidth of frequency components of partial discharge pulses are proposed and many studies are carried out. The HFPD measurement method is able to detect partial discharge signal in active line, therefore this method has a merit that can does condition monitoring of power equipments and measure growth state of partial discharge erosion with on-line and real time. In this study, the radiated HFPD signals are detected by antenna and aging diagnosis is executed with a neural network algorithm. As a result of this study, the aging diagnosis of insulation material using neural networks is possible.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.9
• /
• pp.3357-3376
• /
• 2015

Wireless sensor network is an important research topic that has attracted a lot of attention in recent years. However, most of the interest has focused on wireless sensor network to gather scalar data such as temperature, humidity and vibration. Scalar data are insufficient for diverse applications such as video surveillance, target recognition and traffic monitoring. However, if we use camera sensors in wireless sensor network to collect video data which are vast in information, they can provide important visual information. Video sensor networks continue to gain interest due to their ability to collect video information for a wide range of applications in the past few years. However, how to efficiently store the massive data that reflect environmental state of different times in video sensor network and how to quickly search interested information from them are challenging issues in current research, especially when the sensor network environment is complicated. Therefore, in this paper, we propose a fast algorithm for extracting key frames from video and describe the design and implementation of key frame extraction and sharing in Android for wireless video sensor network.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.9
• /
• pp.3685-3700
• /
• 2015

To avoid influences caused by pose, illumination and facial expression variations, we propose a robust three-step algorithm based on ASM and POEM for facial landmark localization. Firstly, Model Selection Factor is utilized to achieve a pose-free initialized shape. Then, we use the global shape model of ASM to describe the whole face and the texture model POEM to adjust the position of each landmark. Thirdly, a second localization is presented to discriminatively refine the subtle shape variation for some organs and contours. Experiments are conducted in four main face datasets, and the results demonstrate that the proposed method accurately localizes facial landmarks and outperforms other state-of-the-art methods.

• Journal of the Korean Society for Precision Engineering
• /
• v.14 no.10
• /
• pp.135-140
• /
• 1997

The detection of cutting tool states in machining is important for the automation. The information of cutting tool states in metal cutting process is uncertain. Hence a industry needs the system which can detect the cutting tool states in real time and control the feed motion. Cutting signal features must be sifted before the classification. In this paper the Fisher's linear discriminant function was applied to the pattern recognition of the cutting tool states successfully. Cutting conditions and cutting force para- meters have shown to be sensitive to tool states, so these cutting conditions and cutting force paramenters can be used as features for tool state detection.

• Proceedings of the KIEE Conference
• /
• 1988.07a
• /
• pp.723-726
• /
• 1988

This paper is a study about the automatic control of wheelchairs. This is realized by joystick, and is simulated by voice signal recognition. The control system by joystick is designed as follows: joystick paddle is connected with a timer and this timer ouput is high only when the joystick is moved. A computer reads the duration of this high state, and ouputs motor control word decided from this value using look-up table. The control system by voice signal is designed as follows: partial autocorrelation coefficients are computed from A/D converted signals and these values are compared with referance patterns. From this, the motor control word is decided on by the neareast neighbor rule.

• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.1130-1132
• /
• 1997

This paper proposed an application of artificial neural networks to security assessment(SA) in power system. The SA is a important factor in power system operation, but conventional techniques have not achieved the desired speed and accuracy. Since the SA problem involves classification, pattern recognition, prediction, and fast solution, it is well suited for Kohonen neural network application. Self organizing feature map(SOFM) algorithm in this paper provides two dimensional multi maps. The evaluation of this map reveals the significant security features in power system. Multi maps of multi prototype states are proposed for enhancing the versatility of SOFM neural network to various operating state.

• Smart Structures and Systems
• /
• v.6 no.9
• /
• pp.991-1005
• /
• 2010

The key to detecting damage to civil engineering structures is to find an effective damage indicator. The damage indicator should promptly reveal the location of the damage and accurately identify the state of the structure. We propose to use the distance measures of low-order AR models as a novel damage indicator. The AR model has been applied to parameterize dynamical responses, typically the acceleration response. The premise of this approach is that the distance between the models, fitting the dynamical responses from damaged and undamaged structures, may be correlated with the information about the damage, including its location and severity. Distance measures have been widely used in speech recognition. However, they have rarely been applied to civil engineering structures. This research attempts to improve on the distance measures that have been studied so far. The effect of varying the data length, number of parameters, and other factors was carefully studied.

• Journal of the Korea Safety Management & Science
• /
• v.2 no.3
• /
• pp.129-140
• /
• 2000

Nowadays the role of information systems is getting more and more increased according to the development of information technology. In order to manage complex, various and huge data, it is vital to construct efficient information system. For this effective information system, data have properly to be stored, encoded and represented when needed. This paper presents equipment management information system using wireless application protocol. This system enables us to have remote control of data searching and data management. Also, through the technique of data mining, database resulted from this system can be utilized into expectation and analysis about life-cycle, characteristic, and failure time of equipment, pattern recognition of users, and state of movement, etc.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.193-199
• /
• 1992

이 논문에서는 123개 단어의 한국어 음성에 대하여 음성의 대역폭 변화에 따른 인식률을 비교하였다. 인식률 비교실험을 위해 hidden Markov model과 음소와 유사한 131개의 한국어 subword 유니트를 사용한 화자독립 격리단어 인식 시스팀을 사용하였다. 이 실험은 대역폭이 각각 0 - 4.5kHz 및 0.3 - 3.3kHz인 두가지 종류의 음성 데이타베이스를 사용하였다. 훈련과정에서 corrective training의 반복회수를 2로 하고 state transition duration 정보를 사용하였을 때, 0 - 4.5kHz 와 0.3 - 3.3kHz 대역폭에 대해 각각 98.8 % 및 98.2 % 의 최고 인식률을 얻었다. 이로부터 전화대역폭에서도 음성인식률은 크게 저하되지 않음을 알 수 있다.