• Journal of Biomedical Engineering Research
• /
• v.31 no.2
• /
• pp.134-140
• /
• 2010

Readily available high resolution brain MRI scans allow detailed visualization of the brain structures. Researchers have focused on developing methods to quantify shape differences specific to diseased scans. We have developed a novel method to quantify shape information for a specific population based on Multidimensional scaling(MDS). MDS is a well known tool in statistics and here we apply this classical tool to quantify shape change. Distance measures are required in MDS which are computed from pair-wise image registrations of the training set. Registration step establishes spatial correspondence among scans so that they can be compared in the same spatial framework. One benefit of our method is that it is quite robust to errors in registrations. Applying our method to 13 brain MRI showed clear separation between normal and diseased (Cushing's syndrome). Intentionally perturbing the image registration results did not significantly affect the separability of two clusters. We have developed a novel method to quantify shape based on MDS, which is robust to image mis-registration.

• Journal of Navigation and Port Research
• /
• v.34 no.2
• /
• pp.117-122
• /
• 2010

Since fuzzy controllers are nonlinear, it is more difficult to set the controller gains and to analyse the stability compared to conventional PID controllers. This paper proposes a fuzzy PD+I controller for tracking control which uses a linear fuzzy inference(product-sum-gravity) method based on a conventional linear PID controller. In this scheme the fuzzy PD+I controller works similar to the control performance as the linear PD plus I(PD+I) controller. Thus it is possible to analyse and design an fuzzy PD+I controller for given systems based on a linear fuzzy PD controller. The scaling factors tuning scheme, another topic of fuzzy controller design procedure, is also introduced in order to fine performance of the fuzzy PD+I controller. The scaling factors are adjusted by a real-coded genetic algorithm(RCGA) in off-line. The simulation results show the effectiveness of the proposed fuzzy PD+I controller for tracking control problems by comparing with the conventional PID controllers.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.2 no.1
• /
• pp.35-40
• /
• 2004

Final disposal of radioactive waste generated from Nuclear Power Plant (NPP) requires the detailed information about the characteristics and the quantities of radionuclides in waste package. Most of these radionuclides are difficult to measure and expensive to assay. Thus it is suggested to the indirect method by which the concentration of the Difficult-to-Measure (DTM) nuclide is estimated using the correlations of concentration - it is called the scaling factor - between Easy-to-Measure (Key) nuclides and DTM nuclides with the measured concentration of the Key nuclide. In general, the scaling factor is determined by the log mean average (LMA) method and the regression method. However, these methods are inadequate to apply to fission product nuclides and some activation product nuclides such as 14$^{C}$ and 90$^{Sr}$ . In this study, the artificial neural network (ANN) method is suggested to improve the conventional SF determination methods - the LMA method and the regression method. The root mean squared errors (RMSE) of the ANN models are compared with those of the conventional SF determination models for 14$^{C}$ and 90$^{Sr}$ in two parts divided by a training part and a validation part. The SF determination models are arranged in the order of RMSEs as the following order: ANN model

• Journal of the Korean Society of Safety
• /
• v.22 no.5
• /
• pp.65-70
• /
• 2007

This paper is estimation of structure damage caused by VCE(Vapor Cloud Explosion) in enclosure. As we estimate the influence of damage which occur at gas facility in factory. We can utilize the elementary data of safety distance. In this study, the influence of over-pressure caused by VCE in enclosure was calculated by using the Hopkinson's scaling law and the accident damage was estimated by applying the influence on the adjacent structure into the probit model. As a result of the damage estimation conducted by using the probit model, both the damage possibility of explosion overpressure to structures of 20 meters away and to glass bursting of 80 meters away was nearly zero in open space explosion.

• Proceedings of the KAIS Fall Conference
• /
• 2007.05a
• /
• pp.42-45
• /
• 2007

This paper is estimation of structure damage caused by VCE(Vapor Cloud Explosion) in enclosure. As we estimate the influence of damage which occur at gas facility in factory. We can utilize it the elementary data of safety distance. In this study, the influence of explosion over-pressure caused by VCE in enclosure was calculated by using the Hopkinson's scaling law and the accident damage was estimated by applying the influence on the adjacent structure into the probit model. As a result of the damage estimation conducted by using the probit model, both the damage possibility of explosion overpressure to structures 20 meters away and that of overpressure to glass bursting 80m meters away showed nothing.

• Computers and Concrete
• /
• v.22 no.1
• /
• pp.39-51
• /
• 2018

This study introduces a new algorithm to determine size independent values of fracture energy, fracture toughness, and fracture process zone length in three-point bending specimens with shallow to deep notches. By using the exact beam theory, a concept of equivalent notch length is introduced for specimens with no notches in order to predict the peak loads with acceptable precisions. Moreover, the method considers the variations of fracture process zone length and effects of higher order terms of stress field in each specimen size. In this paper, it was demonstrated that the use of some recently developed size effect laws raises some concerns due to the use of nonlinear regression analysis. By using a comprehensive fracture test data, provided by Hoover and Bazant, the algorithm has been assessed. It could be concluded that the proposed algorithm can facilitate a powerful tool for size effect study of three-point bending specimens with different notch lengths.

• Korean Institute of Interior Design Journal
• /
• v.14 no.5 s.52
• /
• pp.185-192
• /
• 2005

The purpose of this study was to clarify spatial composition of the staffs at nursing homes for the elderly with dementia. For this, the data was taken from the staffs in 2 nursing homes in Tokyo by the questionnaire from December4 to 28 in 2004. Replicated MDS (Multi-dimensional Scaling) was employed for data analysis. The results of the study were following. Staffs perceived nursing home by four spatial compositions; facility spaces as like sanitation room, laundry, bathroom, washroom, washbowl, social interaction spaces as like training room, inner court/balcony, living room, staff related spaces as like room for staff, treatment room, fundamental living spaces as like bedroom, dining room. According to two-dimensional perceptual map, sanitation room was wanted near by laundry, treatment room was wanted near by room for staff, bedroom was wanted near by dining room, and training room, living room were wanted near by inner court/balcony. On the other hand, washbowl and washroom were desired somewhere near by every spaces. Therefore, simple shaped washbowls are desired to install in dining room, living room, hallways. And sanitation room and laundry were desired somewhere remote by dining room, bedroom, and living room. Responding upper results, it is recommended planning for visual and sense of smell isolation in case of having difficulty in floor structure and circulation.

• Journal of Korean society of Dental Hygiene
• /
• v.13 no.3
• /
• pp.403-410
• /
• 2013

Objectives : This study aims to provide the basic data for the maintenance and management of healthy implants through finding out the educational experience of toothbrushing and the oral hygiene management for the implant patients and to determine the maintenance measure. Methods : From April 1 to october 1, 2012, 250 patients in the dental clinic in Daejeon completed the self-reported questionnaire. Results : As for the educational experience of toothbrushing, 66.4 percent received the training for toothbrushing and 33.6 percent had not received the training for toothbrushing. Those who brushed three times a day accounted for 74.1 percent. Those who received the training for toothbrushing comprised 56.0 percent and took regular check up of every 6 months. Approximately 50.0% received regular scaling service every 6 months. Periodical medical check up comprised 55.6% among those who received toothbrushing education. Approximately 63.9% thought that periodical medical check up is the most important to maintain oral health. Conclusions : There were important factors for oral health. They were tooth-brushing, the utilization of various oral hygienic products, and the awareness of importance of oral hygiene management. The customized education program for the general public should be implemented as regular as possible.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.346-351
• /
• 2023

With the release of numerous open driving datasets, the demand for domain adaptation in perception tasks has increased, particularly when transferring knowledge from rich datasets to novel domains. However, it is difficult to solve the change 1) in the sensor domain caused by heterogeneous LiDAR sensors and 2) in the environmental domain caused by different environmental factors. We overcome domain differences in the semi-supervised setting with 3-stage model parameter training. First, we pre-train the model with the source dataset with object scaling based on statistics of the object size. Then we fine-tine the partially frozen model weights with copy-and-paste augmentation. The 3D points in the box labels are copied from one scene and pasted to the other scenes. Finally, we use the knowledge distillation method to update the student network with a moving average from the teacher network along with a self-training method with pseudo labels. Test-Time Augmentation with varying z values is employed to predict the final results. Our method achieved 3rd place in ECCV 2022 workshop on the 3D Perception for Autonomous Driving challenge.

• Journal of Communications and Networks
• /
• v.16 no.1
• /
• pp.18-25
• /
• 2014

In this paper, we propose a codebook-based interference alignment (IA) scheme in the constant multiple-input multiple-output (MIMO) interference channel especially for the uplink scenario. In our proposed scheme, we assume cooperation among base stations (BSs) through reliable backhaul links so that global channel knowledge is available for all BSs, which enables BS to compute he transmit precoder and inform its quantized index to the associated user via limited rate feedback link. We present an upper bound on the rate loss of the proposed scheme and derive the scaling law of the feedback load to maintain a constant rate loss relative to IA with perfect channel knowledge. Considering the impact of overhead due to training, cooperation, and feedback, we address the effective degrees of freedom (DOF) of the proposed scheme and derive the maximization of the effective DOF. From simulation results, we verify our analysis on the scaling law to preserve the multiplexing gain and confirm that the proposed scheme is more effective than the conventional IA scheme in terms of the effective DOF.