• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2946-2952
• /
• 2014

Among the Example based Super Resolution(SR) techniques, Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the Learning Sparse-Neighbor Image Representation baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we first use bicubic interpolation to synthesize its high resolution version. We extract the patches from this synthesized image and determine whether each patch corresponds to regions with high or low spatial frequencies. After the weight of each patch is obtained by our method, we used to learn separate SVR models. Finally, we update the pixel values using the previously learned SVRs. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Journal of Convergence for Information Technology
• /
• v.10 no.8
• /
• pp.23-34
• /
• 2020

Recently, the artificial intelligence deep learning field has been hard to commercialize due to the high computing power and the price problem of computing resources. In this paper, we apply a double pruning techniques to evaluate the performance of the in-depth neural network and various datasets. Double pruning combines basic Network-slimming and Parameter-prunning. Our proposed technique has the advantage of reducing the parameters that are not important to the existing learning and improving the speed without compromising the learning accuracy. After training various datasets, the pruning ratio was increased to reduce the size of the model.We confirmed that MobileNet-V3 showed the highest performance as a result of NetScore performance analysis. We confirmed that the performance after pruning was the highest in MobileNet-V3 consisting of depthwise seperable convolution neural networks in the Cifar 10 dataset, and VGGNet and ResNet in traditional convolutional neural networks also increased significantly.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.8 no.1
• /
• pp.33-40
• /
• 2014

The objective of this study was to assess leg muscle activities during whole body vibration (WBV) training with various squat position exercises. Subjects performed four different squat exercises on a vibration platform (VM-10, Sonicworld Co.) Ten healthy women volunteers (age $21{\pm}1.3years$, height $160{\pm}2.1cm$, weight $52{\pm}4.6kg$) were selected. EMG signals four leg muscles were taken and analyzed. The exercises were performed both with 10 seconds WBV and without 10 seconds WBV. Muscle activities during WBV were compared with non-vibration stimulus. In the result, EMG signals caused by WBV were significantly higher (P<0.05) compared with non-vibration stimulus in all leg muscles and squat exercises. The increase in muscle activity caused by Whole body vibration was significantly higher (P<0.05) in LS and OS compared with HU&HS.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.4 s.36
• /
• pp.385-396
• /
• 2005

In this paper, a new structural approach to on-line signature verification is presented. A primitive pattern is defined as a part segmented by a local minimal position of speed. And a structural description of signature is composed of subpatterns which are defined as such forms as rotation shape, cusp shape and bell shape, acquired by composition of the primitives regarding the directional changes. As the matching method to find identical parts between two signatures, a modified DP(dynamic programming) matching algorithm is presented. And also, variation and complexity of local parts are computed from the training samples, and reference model and decision boundary are derived from these. Error rate, execution time and memory usage are compared among the functional approach, the parametric approach and the proposed structural approach. It is found that the average error rate can be reduced from 14.2% to 4.05% when the local parts of a signature are weighted and the complexity is used as a factor of decision threshold. Though the error rate is similar to that of functional approaches. time consumption and memory usage of the proposed structural approach are shown to be very effective.

• Journal of Sasang Constitutional Medicine
• /
• v.27 no.2
• /
• pp.211-221
• /
• 2015

Objectives The aim of this study is to develop a Korea Sasang Constitutional Diagnostic Questionnaire (KS-15) with practicality and validity. Methods The data of body shape, temperament and symptoms were collected from 11 nationwide multi-centers in Korean Medical Clinics between Nov. 2009 to Aug. 2012. KS-15 algorithm was developed through Multinomial Logistic Regression and was applied for test group to verify it's classificatory function. Results The weight value of variables of KS-15 were different between male and female. The concordance of KS-15 with the expert in training group was 58.34% (72.8% in Tae-eumin, 54.8% in Soeumin, and 44.1% in Soyangin). The concordance of KS-15 with the expert in test group was 63.1% (63% in Tae-eumin, 83.3% in Soeumin, and 51.9% in Soyangin). Conclusions KS-15 seems to have suitable implement with practicality and validity. Further study is necessary to verify it accuracy and cut-off.

• Journal of Life Science
• /
• v.20 no.7
• /
• pp.1100-1106
• /
• 2010

The purpose of this study was to analyze the effects of Hatha Yoga exercise on serum leptin and metabolic syndrome factors in obese and menopausal middle-aged women. The subjects were 26 obese women divided into the Hatha Yoga exercise group (n=13), which trained for 16 weeks, and the Control group (n=13). Variables of body composition, serum leptin and metabolic syndrome factors were measured in all the subjects before and after the 16-week Hatha Yoga training. The results of the study in the Hatha Yoga group were as follows: body weight, % fat, BMI, WC, WHR and VFA had significantly decreased, but SMM had increased. HDL-C had significantly increased, but leptin, TC, TG, LDL-C, insulin, glucose and HOMA-IR had decreased. The main variables affecting changes in VFA were % body fat, BMI, WHR, TC, LDL-C, glucose, and HOMA-IR. Therefore, regular and continuous Hatha Yoga exercise was effective in improving body composition, visceral fat and serum lipids. Consequently, Hatha Yoga exercise will be effective in preventing cardiovascular disease caused by obesity in obesew and menopausal middle-age women.

• Physical Therapy Korea
• /
• v.14 no.3
• /
• pp.41-47
• /
• 2007

Dance movements consist of combinations of movements such as jumping, rotation, maintaining balance, leg lifts, and plantar flexion with toe shoes. Dance movements require great muscle strength of lower extremities as well as muscular endurance. The purposes of this study were to investigate correlation between the anthropometric variables and the peak torque in young female dancers and to identify variables that affect isokinetic peak torque. Twenty-six female dancers ($19.7{\pm}1.2$ years of age) performed concentric maximum force efforts on the knee extensors and flexors at $60^{\circ}/sec$ and $120^{\circ}/sec$, the ankle plantar flexors (PF) and dorsiflexors (DF) at $30^{\circ}/sec$ and $120^{\circ}/sec$. Antropometric variables such as age, height, weight, body mass index (BMI), thigh girth, calf girth and duration of dance training were measured. To identify antropometric variables related to muscle strength, Pearson correlations were computed and a stepwise multiple regression analysis was performed. Pearson correlation coefficients of Knee extensor at $60^{\circ}/sec$ and $120^{\circ}/sec$ revealed moderate positive associations with BMI and thigh girth. Pearson correlation coefficients of ankle PF revealed low-to-moderate positive associations with height. Ankle DF also revealed moderate positive associations with BMI and calf girth. The main predictor variables of knee extensor at 60/sec, were stepwise multiple regression, age, height, thigh and girth.

• Preventive Nutrition and Food Science
• /
• v.10 no.2
• /
• pp.151-155
• /
• 2005

This study was intended to investigate the effects of regular swimming exercise and vitamin C supplementation on the antioxidant system following exercise stress. For the swimming exercise experiment, a swimming adaptation exercise of 1 week was given to a group of 6-week-old mice. Following this, a swimming exercise for 8 weeks was conducted. The experimental group was divided into 3: a control group (C), a swimming exercise trained group (T), and a group of swimming + vitamin C supplementation (TC: vitamin supplementation: 1.3 mg/l00 g diet). After the swimming exercise, these group were further divided into those that had received the exercise stress for 2 hours and those that had not experienced exercise stress group. Then, the activities of the superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and malondialdehyde (MDA) concentrations were measured. There was a lower weight increase in the T and TC groups than in the C group, and there was no significant difference between T and TC group. When exercise stress was not experienced, the activity of SOD was significantly increased in the TC group than in the T group, but there was no significant difference between C and T groups. The groups that had experienced a 2-hour exercise stress showed the SOD activity levels according to the following order, C < T < TC, with a significant difference between the three groups (p<0.05). There was no difference in MDA concentration amongst the experimental groups in non-exercise stress group. As well, there was no differences in MDA concentration between the C group and T group in the 2 hour exercise stress group. However, the TC group showed a MDA concentration level significantly lower than that of the T group. A significant increase in MDA concentration was observed in C group, when exercise stress was provided with no significant difference in the T and TC groups. As a result, regular exercise and vitamin C supplementation can be considered important in controlling the formation of lipid peroxides in exercise stress.

• Journal of Intelligence and Information Systems
• /
• v.16 no.2
• /
• pp.95-108
• /
• 2010

In this paper, we present a hybrid neural network model for dynamic hand gesture recognition. The model consists of two modules, feature extraction module and pattern classification module. We first propose a modified CNN(convolutional Neural Network) a pattern recognition model for the feature extraction module. Then we introduce a weighted fuzzy min-max(WFMM) neural network for the pattern classification module. The data representation proposed in this research is a spatiotemporal template which is based on the motion information of the target object. To minimize the influence caused by the spatial and temporal variation of the feature points, we extend the receptive field of the CNN model to a three-dimensional structure. We discuss the learning capability of the WFMM neural networks in which the weight concept is added to represent the frequency factor in training pattern set. The model can overcome the performance degradation which may be caused by the hyperbox contraction process of conventional FMM neural networks. From the experimental results of human action recognition and dynamic hand gesture recognition for remote-control electric home appliances, the validity of the proposed models is discussed.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.9
• /
• pp.5789-5794
• /
• 2015

Recently, thin and light-weight production technologies are needed in IT industry in accordance with increase of the smart phones and mobile PC products. In order to make light and high rigidity products, engineering plastic and aluminum materials are frequently used in products appearance and frame hat support structure. Especially aluminum extrusion and CNC Brick processes are widely used for high strength and high rigidity technology. But extrusion method has constraints to apply exterior design and CNC Brick process has relatively high production cost and low speed of manufacturing. In this research, a new process method is introduced in order to reduce material cost and to improve manufacturing speed dramatically. Plate forging process means basically that thickening of local wall area thickness after deform exterior shape by deep drawing and bending process. Therefore, it is possible to minimize the waste of material and the manufacturing time. In this study the process of plate forging is designed using finite element program AFDEX-2D and the thickness and the width of initial deformed blank. And it is verified as a sample which is a part of laptop developed through the proposed plate forging method.