• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.6
• /
• pp.681-686
• /
• 2012

This paper studied the speech parameters less affected by the human emotion for the development of the robust emotional speech recognition system. For this purpose, the effect of emotion on the speech recognition system and robust speech parameters of speech recognition system were studied using speech database containing various emotions. In this study, mel-cepstral coefficient, delta-cepstral coefficient, RASTA mel-cepstral coefficient, root-cepstral coefficient, PLP coefficient and frequency warped mel-cepstral coefficient in the vocal tract length normalization method were used as feature parameters. And CMS (Cepstral Mean Subtraction) and SBR(Signal Bias Removal) method were used as a signal bias removal technique. Experimental results showed that the HMM based speaker independent word recognizer using frequency warped RASTA mel-cepstral coefficient in the vocal tract length normalized method, its derivatives and CMS as a signal bias removal showed the best performance.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.5
• /
• pp.13-20
• /
• 2019

Recently, object detection technology based on CNN has been actively studied. Object detection technology is used as an important technology in autonomous vehicles, intelligent image analysis, and so on. In this paper, we propose a rotation change robust object detection system by applying MI-FL (Moment Invariant-Feature Layer) to SSD (Single Shot Multibox Detector) which is one of CNN-based object detectors. First, the features of the input image are extracted based on the VGG network. Then, a total of six feature layers are applied to generate bounding boxes by predicting the location and type of object. We then use the NMS algorithm to get the bounding box that is the most likely object. Once an object bounding box has been determined, the invariant moment feature of the corresponding region is extracted using MI-FL, and stored and learned in advance. In the detection process, it is possible to detect the rotated image more robust than the conventional method by using the previously stored moment invariant feature information. The performance improvement of about 4 ~ 5% was confirmed by comparing SSD with existing SSD and MI-FL.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.4
• /
• pp.75-80
• /
• 2021

This paper relates with the Hybrid-DSE-MMA (Hybrid-Dithered Signed Error-MMA) that is possible to improving the equalization performance by using the adaptive modulus and adaptive stepsize in DSE-MMA adaptive equalizer. The DSE-MMA possible to improve the robustness performance to external noise of SE-MMA by using the sign after adding the dither signal for get the error signal in order to update the tap coefficient. But it has a drawback of performance degradation in convergence speed and residual isi by using the fixed modulus and fixed stepsize. In this paper, it was confirmed that this equalization performance degradation was improved by applying the adaptive modulus and stepsize in DSE-MMA propotional to the output power of equalizer by computer simulation. In order to compare the improved equalization performance to currently DSE-MMA, the recovered signal constellation that is the output of the equalizer, residual isi, Maximum Distortion, MSE and the SER were used as a performance index. As a result of computer simulation, the Hybrid-DSE-MMA improve the equalization performance in every index, but gives slower convergence speed compared to DSE-MMA.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.02a
• /
• pp.211-212
• /
• 2011

플라즈마를 제어하기 위해서는 플라즈마의 온도, 밀도, 에너지 분포등과 같은 플라즈마의 특성을 정확히 측정할 수 있어야한다. 핵융합발전에서는 플라즈마를 발생하기 위하여 플라즈마의 온도, 밀도 등 각종 변수들을 시공간적으로 계측, 분석할 수 있는 진달설비를 사용하고 있으며, 정확한 플라즈마 제어와 측정을 위한 새로운 진단기술을 개발하고 있다. 그리고 중요한 변수중에 하나인 플라즈마 이온온도를 측정하기 위해 중성입자 검출법이 잘 알려져 있다. 이 실험은 수소 중성입자가 토카막 내부의 플라즈마 이온과 충돌하면서 생성된 고속 중성입자의 에너지를 분석하는 실험이다. 본 연구의 실험방법은 수소 중성입자를 이온빔 장치에서 이온화 시킨 후 자체 제작한 가속기를 통하여 가속시켜 에너지 특성을 분석을 하는 것이다. 본 연구의 실험장치로 에너지 교정용 100 keV 이온빔 소스를 제작 하였고 이온빔 장치 내부에 수소기체를 주입하고 기체방전을 일으켜 플라즈마를 발생시켰다. 이온빔 외부에는 팬을 설치하고 전도성이 강한 물 대신 전도성이 약한 오일을 사용하여 냉각 하였다. 이온빔 장치와 결합될 이온 가속장치는 지름 300 mm, 두께 2 mm의 원형 구리판을 여러층으로 쌓아 전극으로 제작하였고 전극과 전극 사이에서 코로나 방전과 스파크를 방지하기 위해 전극 둘레에 코로나링을 설치 하였다. 또한 전극 사이마다 1G${\Omega}$의 저항을 설치한 후 고전압을 생성하여 이온 가속 효율을 증대시켰다. 진공시스템으로는 Alcatel사의 CFF100 터보분자 펌프와 우성진공사의 MVP24 진공로타리펌프를 결합하여 사용하였으며, 진공도측정은 Alcatel사의 ACS1000 장치를 사용하였다. 고진공후 고속 중성입자의 이온화와 에너지 측정을 위한 전하교환기를 설치하였다. 전하교환기로는 진공시스템을 별도로 설치하고 비용이 비교적 많이 드는 기체형 전하교환기 대신 소형화가 가능하고 유지보수가 좋은 고체형 전하교환기 제작하여 실험 하였다. 전하교환기에서 이온화된 고속 중성입자가 전기장이나 자장에 영향을 받았을때 에너지분포를 디텍터를 통해 측정하였다. 즉, 이온화된 중성입자의 에너지가 실리콘 다이오드를 통해 전압 펄스 신호로 변환되고 이차 증폭기를 통해 전압 펄스 신호들이 증폭한다. 에너지 측정을 위한 디텍터는 소형화가 가능하고 비용이 비교적 적게 드는 실리콘 다이오드를 설치하였다. 본 연구결과 중성입자 에너지 분석 장치가 실제 핵융합 장치의 플라즈마 이온온도와 특성 측정에 적용할 수 있으며, 앞으로 개발될 여러 형태의 응용 플라즈마 발생장치의 플라즈마 진단에 이용될 것으로 기대한다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.39 no.1
• /
• pp.55-61
• /
• 2019

To improve bus service quality and passengers' safety at bus stops, it is necessary to analyze the relationship between bus stopping behaviors and bus stop conditions. Thus, this study investigated the proper stopping rate at 31 bus stops on Central Boulevard in Busan metropolitan city and analyzed the relationship between the proper stopping rate and bus stop operating conditions such as the length of berth, the distance from the front line of berth to bus shelter and so on. The analysis results show that the proper stopping rate is 45.82 % and the rates are closely related to each bus stop operating condition. Therefore, it is highly recommended that the front line of berth should be located parallel to bus shelter for improving the safety of bus passengers. In addition, it is suggested that the length of berth should be longer than 15m and the pavement marking of berth be redesigned with dash lines to take account of buses' entering and leaving the berth.

• Journal of the Korea Convergence Society
• /
• v.7 no.5
• /
• pp.227-235
• /
• 2016

This research focuses on the life history of university professor and the meaning of their lives to provide career information to parents and teachers conducting childhood career education. professor K with 25 years of professorship was targeted for insides's view(emic), which describes the life story through the professor's voice and words, was used as the method of research and additional data was collected through in-depth interviews with professor K, his writings, e-mails and in-depth interviews. The collected data was electronically transferred and categorized for field and categorical analysis. The analysis showed that the motive for professor K to enter academic society was due to his utmost efforts, assistance of family, and his friend encountering. Also after entering the university, he served as key role in research, education and as an appointed professor. After retirement, he is contributing to the society as a expert and practitioner. As such, 'continuous effort and consistent personal innovation', 'strong belief and clear calling', 'warm humanity and practical life', and 'flexible attitude and educational passion' were deducted from professor K 's professional capability.

• Journal of Internet Computing and Services
• /
• v.17 no.5
• /
• pp.17-23
• /
• 2016

Wireless sensor-actuator networks (WSANs) enhance the existing wireless sensor networks (WSNs) by equipping sensor nodes with an actuator. The actuators work with the sensor nodes and perform application-specific operations. The WSAN systems have several applications such as disaster relief, intelligent building, military surveillance, health monitoring, and infrastructure security. These applications require capability of reliable data transfer to act responsively and accurately. Biologically inspired modeling techniques have received considerable attention for achieving robustness, scalability, and adaptability, while retaining individual simplicity. In this paper, an epidemic-inspired algorithm for data dissemination with delay constraints while minimizing energy consumption in WSAN is proposed. The steady states and system stability are analyzed using control theory. Also, simulation results indicate that the proposed scheme provides desirable dissemination delay and energy saving.

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.4
• /
• pp.274-282
• /
• 2009

Various location tracking sensors; such as GPS, INS, radar, and optical equipment; are used for tracking moving targets. In order to effectively track moving targets, it is necessary to develop an effective fusion method for these heterogeneous devices. There have been studies in which the estimated values of each sensors were regarded as different models and fused together, considering the different error characteristics of the sensors for the improvement of tracking performance using heterogeneous multi-sensor. However, the rate of errors for the estimated values of other sensors has increased, in that there has been a sharp increase in sensor errors and the attempts to change the estimated sensor values for the Sensor Probability could not be applied in real time. In this study, the Sensor Probability is obtained by comparing the RMSE (Root Mean Square Error) for the difference between the updated and measured values of the Kalman filter for each sensor. The process of substituting the new combined values for the Kalman filter input values for each sensor is excluded. There are improvements in both the real-time application of estimated sensor values, and the tracking performance for the areas in which the sensor performance has rapidly decreased. The proposed algorithm adds the error characteristic of each sensor as a conditional probability value, and ensures greater accuracy by performing the track fusion with the sensors with the most reliable performance. The trajectory of a UAV is generated in an experiment and a performance analysis is conducted with other fusion algorithms.

• Journal of Digital Convergence
• /
• v.18 no.9
• /
• pp.285-295
• /
• 2020

The purpose of this study was to identify factors affecting college-based stress and develop stress-relieving educational programs. Data were collected from 169 students in G-province, from June 15th to july 15th, 2020. Analysis was done using IBM SPSS 21.0 for frequency, percentage, average, standard deviation, t-test, ANOVA, Pearson correlation coefficient, and Multiple regression. College-based stress positively correlated with life event, life event importance, interpersonal emotion, and self-control. The most important factor affecting college-based stress was psychological hardiness (β=-.323, p<.001), followed by club activities (β=.237, p<.001), interpersonal conflict (β=.215, p=.001), grade-2 (β=-.203, p=.001), department adaptation (β=.161, p=.010), and age-21 (β=-.149, p=.024). The explanatory power was 44.4% (F=42.43, p<.001). The results of this study are expected to be used as basic data for the development of educational programs for stress reduction at the university level. Therefore, educational programs considering age and grade, using club activities as support systems, and dealing with psychological hardiness and department adaptation are needed.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.8
• /
• pp.331-338
• /
• 2022

Successor representation (SR) is a model of human reinforcement learning (RL) mimicking the underlying mechanism of hippocampal cells constructing cognitive maps. SR utilizes these learned features to adaptively respond to the frequent reward changes. In this paper, we evaluated the performance of SR under the context where changes in latent variables of environments trigger the reward structure changes. For a benchmark test, we adopted SR-Dyna, an integration of SR into goal-driven Dyna RL algorithm in the 2-stage Markov Decision Task (MDT) in which we can intentionally manipulate the latent variables - state transition uncertainty and goal-condition. To precisely investigate the characteristics of SR, we conducted the experiments while controlling each latent variable that affects the changes in reward structure. Evaluation results showed that SR-Dyna could learn to respond to the reward changes in relation to the changes in latent variables, but could not learn rapidly in that situation. This brings about the necessity to build more robust RL models that can rapidly learn to respond to the frequent changes in the environment in which latent variables and reward structure change at the same time.