http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
An Efficient Electronic Wheelchair Seat Balancing Maintain Methodology Applying Smart Sensors
Sanghyun Park,Jinsul Kim,Yonggwan Won 보안공학연구지원센터 2015 International Journal of Smart Home Vol.9 No.1
In this paper we propose a new solution which utilizing the Gyro Sensor and Tilt Sensor in real-time controlling the balance of wheelchair seat. The existing wheelchair systems have the risk of falling when users go uphill because of the uneven weight allocation. But when applying our new method, the wheelchair seat can automatically adjust to maintain the balance therefore provides a safer and more stable ride. In order to adjust the seat correctly, we calculate the seat angle using Tilt Sensor. However, due to the law of inertia when wheelchair is moving there is deviation in output value of Tilt Sensor. To solve this problem, a Gyro sensor is used for measuring the acceleration and get the correct value of tilting angle. Through various experimentations we can prove that by taking advantage of Gyro sensor and Tilt sensor combination, the system is able to determine the correct seat angle in both cases when the wheelchair moving or not. We also tackle the power consumption issue in wheelchair, by using ZigBee sensor module to retrieve terrain information in advance and controlling the balance by two motors thus the overall power consumption for seat balancing is reduced significantly.
Two-Stage Extreme Learning Machine for SLFNs in Regression
Hieu Trung Huynh,Yonggwan Won 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
Neural network approach has been massively used in regression problem. However, collected data for training often include outliers which affect the final results. In this paper, we propose a new approach for outlier elimination in regression based on the extreme learning machine (ELM) called two-stage ELM. Training process consists of two stages. In the first stage, the single-hidden layer feed forward neural network (SLFN) is trained based on the ELM algorithm using the whole of training set. The trained SLFN is used to verify training patterns. Patterns corresponding to the outputs exceeding a rejection threshold are removed from the training set. Finally, the remainder of training set is used to train the SLFN again based on the ELM algorithm. One interesting observation is that, our approach is simple to detect and eliminate outliers and it can be able to deliver lower error than that of the normal ELM with fast learning speed if there exist outliers in the training set.
Hematocrit estimation using online sequential extreme learning machine.
Huynh, Hieu Trung,Won, Yonggwan,Kim, Jinsul Pergamon Press 2015 Bio-medical materials and engineering Vol.26 No.1
<P>Hematocrit is a blood test that is defined as the volume percentage of red blood cells in the whole blood. It is one of the important indicators for clinical decision making and the most effective factor in glucose measurement using handheld devices. In this paper, a method for hematocrit estimation that is based upon the transduced current curve and the neural network is presented. The salient points of this method are that (1) the neural network is trained by the online sequential extreme learning machine (OS-ELM) in which the devices can be still trained with new samples during the using process and (2) the extended features are used to reduce the number of current points which can save the battery power of devices and speed up the measurement process.</P>
Extension of General Convergence Framework with Significant Samples
Nguyen Ha Vo,Yonggwan Won 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
Single Class Classification is the problem of distinguishing one class of data (called positive class) from the universal set of multiple classes (negative class). In this paper, we proposed an improvement of Extended General Mapping Convergence framework using extreme learning machine, a recently developed machine learning algorithm. This proposed method keeping the high accuracy in classification while improving the high speed of old method.
Performance Enhancement of SLFNs in Classification by Reducing Effect of Outliers
Hieu Trung Huynh,Yonggwan Won 대한전자공학회 2007 ITC-CSCC :International Technical Conference on Ci Vol.2007 No.7
In this paper, an approach for outlier reduction is proposed to enhance the classification performance of the single-hidden layer feed-forward neural networks (SLFNs). Outliers in the data set are detected based on the distribution on every feature, in which scores are assigned to patterns. Patterns detected as outliers based on these scores will contribute very little in estimating the weights of SLFNs. The experimental results show that, the proposed approach can obtain high accuracy with fast learning speed if there exist outliers in the training set.