• Journal of Electrical Engineering and Technology
• /
• v.4 no.2
• /
• pp.219-228
• /
• 2009

The growing application and the numerous qualities of induction motors (1M) in industrial processes that require high security and reliability levels has led to the development of multiple methods for early fault detection. However, various faults can occur, such as stator short-circuits and rotor failures. Traditionally the diagnosis machine is done through a sinusoidal power supply, in the present paper we study experimentally the effects of the rotor failures, such as broken rotor bars in function of the ac supplying, the load and show the impact of the converter from diagnosis of the machine. The technique diagnosis used is based on the spectral analysis of stator currents or stator voltages respectively according to the types of induction motor ac supplying. So, four different ac supplying are considered: ${\odot}$ the IM is directly by the balanced three-phase network voltage source, ${\odot}$ the IM is fed by a sinusoidal current source given the controlled by hysteresis, ${\odot}$ the IM is fed (in open loop) by a scalar control imposing through ratio V/f=constant, ${\odot}$ the IM is controlled through a vector control using space vector pulse width modulation (SVPWM) technique inverter with an outer speed loop.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.251-258
• /
• 2003

A support vector machine (SVM) is supposed to provide a good generalization performance, but the actual performance of a actually implemented SVM is often far from the theoretically expected level. This is largely because the implementation is based on an approximated algorithm, due to the high complexity of time and space. To improve this limitation, we propose ensemble of SVMs by using Bagging (bootstrap aggregating) and Boosting. By a Bagging stage each individual SVM is trained independently using randomly chosen training samples via a bootstrap technique. By a Boosting stage an individual SVM is trained by choosing training samples according to their probability distribution. The probability distribution is updated by the error of independent classifiers, and the process is iterated. After the training stage, they are aggregated to make a collective decision in several ways, such ai majority voting, the LSE(least squares estimation) -based weighting, and double layer hierarchical combining. The simulation results for IRIS data classification, the hand-written digit recognition and Face detection show that the proposed SVM ensembles greatly outperforms a single SVM in terms of classification accuracy.

• Journal of Internet Computing and Services
• /
• v.5 no.6
• /
• pp.79-91
• /
• 2004

Various facial pose detection and recognition has been a difficult problem. The problem is due to the fact that the distribution of various poses in a feature space is mere dispersed and more complicated than that of frontal faces, This thesis proposes a robust pose-expression-invariant face recognition method in order to overcome insufficiency of the existing face recognition system. First, we apply the TSL color model for detecting facial region and estimate the direction of face using facial features. The estimated pose vector is decomposed into X-V-Z axes, Second, the input face is mapped by deformable template using this vectors and 3D CANDIDE face model. Final. the mapped face is transformed to frontal face which appropriates for face recognition by the estimated pose vector. Through the experiments, we come to validate the application of face detection model and the method for estimating facial poses, Moreover, the tests show that recognition rate is greatly boosted through the normalization of the poses and expressions.

• Steel and Composite Structures
• /
• v.19 no.3
• /
• pp.569-583
• /
• 2015

Recent years, explosive welding structures have been widely used in many engineering fields. The bonding state detection of explosive welding structures is significant to prevent unscheduled failures and even catastrophic accidents. However, this task still faces challenges due to the complexity of the bonding interface. In this paper, a new method called dual-tree complex wavelet transform based permutation entropy (DTCWT-PE) is proposed to detect bonding state of such structures. Benefiting from the complex analytical wavelet function, the dual-tree complex wavelet transform (DTCWT) has better shift invariance and reduced spectral aliasing compared with the traditional wavelet transform. All those characters are good for characterizing the vibration response signals. Furthermore, as a statistical measure, permutation entropy (PE) quantifies the complexity of non-stationary signals through phase space reconstruction, and thus it can be used as a viable tool to detect the change of bonding state. In order to more accurate identification and detection of bonding state, PE values derived from DTCWT coefficients are proposed to extract the state information from the vibration response signal of explosive welding structure, and then the extracted PE values serve as input vectors of support vector machine (SVM) to identify the bonding state of the structure. The experiments on bonding state detection of explosive welding pipes are presented to illustrate the feasibility and effectiveness of the proposed method.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2004.05b
• /
• pp.603-607
• /
• 2004

The vision-based driver fatigue detection is one of the most prospective commercial applications of facial expression recognition technology. The facial feature tracking is the primary technique issue in it. Current facial tracking technology faces three challenges: (1) detection failure of some or all of features due to a variety of lighting conditions and head motions; (2) multiple and non-rigid object tracking and (3) features occlusion when the head is in oblique angles. In this paper, we propose a new active approach. First, the active IR sensor is used to robustly detect pupils under variable lighting conditions. The detected pupils are then used to predict the head motion. Furthermore, face movement is assumed to be locally smooth so that a facial feature can be tracked with a Kalman filter. The simultaneous use of the pupil constraint and the Kalman filtering greatly increases the prediction accuracy for each feature position. Feature detection is accomplished in the Gabor space with respect to the vicinity of predicted location. Local graphs consisting of identified features are extracted and used to capture the spatial relationship among detected features. Finally, a graph-based reliability propagation is proposed to tackle the occlusion problem and verify the tracking results. The experimental results show validity of our active approach to real-life facial tracking under variable lighting conditions, head orientations, and facial expressions.

• The Journal of the Korea Contents Association
• /
• v.7 no.3
• /
• pp.93-100
• /
• 2007

Internet had spread in all fields with the fast speed during the last 10 years. Lately, wireless network is also spreading rapidly. Also, number of times that succeed attack attempt and invasion for wireless network is increasing rapidly TMS system was developed to overcome these threat on wireless network. Existing TMS system supplies active confrontation mechanism on these threats. However, existent TMS has limitation that new form of attack do not filtered efficiently. Therefor this paper proposes a new method that it automatically compute the threat from the imput packets with vector space model and detect anomaly detection of wireless network. Proposed mechanism in this research analyzes similarity degree between packets, and detect something wrong symptom of wireless network and then classify these threats automatically.

• Design & Manufacturing
• /
• v.17 no.3
• /
• pp.34-40
• /
• 2023

In this study, real-time collection of mold vibration signals during injection molding processes was achieved through IoT devices installed on the mold surface. To analyze changes in the collected vibration signals, injection molding was performed under six different process conditions. Analysis of the mold vibration signals according to process conditions revealed distinct trends and patterns. Based on this result, cosine similarity was applied to compare pattern changes in the mold vibration signals. The similarity in time and acceleration vector space between the collected data was analyzed. The results showed that under identical conditions for all six process settings, the cosine similarity remained around 0.92±0.07. However, when different process conditions were applied, the cosine similarity decreased to the range of 0.47±0.07. Based on these results, a cosine similarity threshold of 0.60~0.70 was established. When applied to the analysis of mold vibration signals, it was possible to determine whether the molding process was stable or whether variations had occurred due to changes in process conditions. This establishes the potential use of cosine similarity based on mold vibration signals in future applications for real-time monitoring of molding process changes and anomaly detection.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2006.05a
• /
• pp.1198-1206
• /
• 2006

프로그램 소스코드 표절 검사에 대한 기존 방법은 크게 지문(finger-print)법과 구조기반 검사법으로 나뉘며, 주로 단어의 유사성이나 발생빈도를 사용하거나 소스코드 구조상의 특징으로 두 소스간의 유사성을 비교한다. 본 연구에서는 프로그래밍 언어의 예약어 시퀀스를 사용하여 소스코드들 간의 유사성을 비교하고, 이 결과를 FCA(Formal Concept Analysis)를 통해 해석하고 시각화 하는 방법을 제시한다. 일반적인 VSM(Vector Space Model)과 같은 단일 단어 분석으로는 단어의 인접성을 구분할 수 없으므로 단어의 시퀀스 분석이 가능하도록 알고리즘을 구성하였으며 이러한 방식은 지문법의 단점인 소스코드의 부분적인 표절 탐지의 난점을 해결할 수 있고 함수의 호출 순서나 수행 순서에 상관없이 표절을 탐지할 수 있는 장점을 가진다. 마지막으로 유사도 측정결과는 FCA를 이용하여 격자(lattice)로 시각화됨으로써 이용자의 이해도를 높일 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.478-480
• /
• 2001

본 논문에서는 실제 도로 영상에서 움직이는 물체를 검출하고 판별하기 위한 새로운 방법을 제안한다. 제안된 방법은 연속된 영상의 차영상에 적응적 임계간을 적용하여 움직임이 있는 후보 영역을 검출한다. 검출된 후보영역에 관심의 대상이 되는 물체의 포함 여부를 판별하기 위해 신경망을 사용한다. 신경망의 입력으로 사용되는 특징 벡터들의 차원을 줄이기 위해, 후보 영역의 스케일 공간 웨이블릿 특징 벡터 (scale-space wavelet feature vector)들을 사용한다. 제안된 방법은 비디오 기반의 응용 프로그램에 유용하게 이용될 수 있으며 특히, 시간에 따라 조명이 변하거나 잡음이 포함된 비디오 영상에 대해 좋은 결과를 얻을 수 있다.

• Proceedings of the KIPE Conference
• /
• 1999.07a
• /
• pp.274-277
• /
• 1999

The paper describes a implementation of space vector pulse-width modulation (SVPWM) voltage source inverter using Field Programmable Gate Array(FPGA) for a induction motor control system. The implemented chip is included logic circuits for SVPWM, dead time compensation and speed detection using Quick Logic, QL16X24B. The maximum operating frequency and delay time can be set to 110MHz and 6 nsec. The designed FPGA for SVPWM can be incorporated with a digital signal processing to provide a simple and effective solution for high performance voltage source inverter drives. Simulation and Implementation results are shown to verify the usefulness of FPGA as a Application Specific Integrated Circuit(ASIC) in power electronics applications