• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.12
• /
• pp.1682-1687
• /
• 2020

We study the problem of sampling a subset of nodes of graphs for bandlimited graph signals such that the signal values on the sampled nodes provide the most information in order to reconstruct the original graph signal. Instead of directly minimizing the reconstruction error, we focus on minimizing the upper bound of the reconstruction error to reduce the complexity of the selection process. We further simplify the upper bound by applying useful approximations to propose a low-weight greedy selection process that is iteratively conducted to find a suboptimal sampling set. Through the extensive experiments for various graphs, we inspect the performance of the proposed algorithm by comparing with different sampling set selection methods and show that the proposed technique runs fast while preserving a competitive reconstruction performance, yielding a practical solution to real-time applications.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.22 no.4
• /
• pp.32-40
• /
• 1986

A model experiment of purse seine by the circulating water tank was carried out on the changes of net shape and the tension of purseline under operation in two layer current. In the circular tank, the two layer current was made by cutting off the current of upper layer and producing the bottom current by the equipment shown in Fig. 1. The model experiment of purse sein was made on a reduced scale 1 :400, and the experiment was carried out according to the Tauti's model law. When the bottom current of O. 5 knot flows to lower part of three-eighths of net, following results are derived. The depth of sinkerline reached only about 80% of that of no current set. The horizontal shift of sinker line caused by the bottom current is maximized in tight set. The enclosed area by the floatIine immediately after the completion of set net is 61. 5% in tight set, 50. 0 % in loose set and 54. 1 % in lateral set of those in the case of no current. In the first half period of pursing, the tension of the purseline is enhenced by the bottom current and the pattern of increasing is irregular in the tension curves.

• Honam Mathematical Journal
• /
• v.32 no.4
• /
• pp.625-632
• /
• 2010

The notions of commutative pre-logics and terminal sections are introduced. Characterizations of a commutative prelogic are provided. Properties of deductive systems in pre-logics which are upper semilattices are considered.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.39-43
• /
• 2000

In this paper, we show that the fuzzy sample mean is a strong consistent estimator for the expectation of a fuzzy random set taking values in the space F(R$\^$/p) of upper semicontinuous convex fuzzy subsets of R$\^$/p with compact support.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.360-365
• /
• 1995

We shall define the usual fuzzy distance between two fuzzy points in R, the set of all real, numbers, using the usual distance between two points in R. Applying the notion of this usual fuzzy distance, we construct the usual fuzzy topology for R, introduce the notions of lower, stationary and upper cover and obtain the fuzzy Heine-Borel theorem.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.2037-2040
• /
• 2004

In this paper we introduce a new prototype based classifiers for overlapping data, where training pattern can be overlap on the feature space. The proposed classifier is based on the prototype from neural network classifier (NNC)[1] for overlap data. The method automatically chooses the initial center and two radiuses for each class. The center is used as a mean representative of training data for each class. The unclassified pattern is classified by measure distance from the class center. If the distance is in the lower (shorter radius) the unknown pattern has the high percentage of being in this class. If the distance is between the lower and upper (further radius), the pattern has the probability of being in this class or others. But if the distance is outside the upper, the pattern is not in this class. We borrow the words upper and lower from the rough set to represent the region of certainty [3]. The training algorithm to find number of cluster and their parameters (center, lower, upper) is presented. The clustering result is tested using patterns from Thai handwritten letter and the clustering result is very similar to human eyes clustering.

• Physical Therapy Korea
• /
• v.11 no.1
• /
• pp.75-83
• /
• 2004

The purpose of this study was to research the effects of vocalization on upper extremity motion during occupational performance and to compare non-meaning and meaning vocalization. Experiments were performed on 30 subjects. They had no medical history of neurological problems with their upper extremities. Using a tea cup, a tea tray, and a tea spoon, they set a table during vocalization. We used meaning and non-meaning vocalization with the subjects. An example of meaning vocalization would be naming something, and an example of non-vocalization would be saying, "Ah." We used a 3-D analysis system called CMS-HS. We analyzed the motion in the angular velocity and acceleration of the elbow while recording performance time. The results of this study showed that vocalization enhanced the angular velocity and acceleration of the elbow, and also enhanced performance time. In short, vocalization improved upper extremity motion by making it faster and smoother. There were no significant differences between meaning and non-meaning vocalization.

• Korean Journal of Applied Biomechanics
• /
• v.31 no.1
• /
• pp.72-78
• /
• 2021

Objective: The purpose of this study was to investigate relations and effectiveness about mountain climbling exercise with different level of support surfaces by analyzing heart rate and EMG data. A total of 10 male college students with no musculoskeltal disorder were recruited for this study. Method: The biomechanical analysis was performed using heart rate monitor (Polar V800, Polar Electro Oy, Finland), step-box, exercise mat, and EMG device (QEMG8, Laxtha Inc. Korea, sampling frequency = 1,024 Hz, gain = 1,000, input impedance > 1012 Ω, CMRR > 100 dB). In this research, step-box were used to create different surface levels on the upper body (flat surface, 10% of subject's height, 20% of subject's height, and 30% of subject's hight). Based on these different conditions, data was collected by performing mountain climbing exercise during 30 seconds. Subjects were given 5 minutes of break to prevent muscular fatigue after each exercise. For each dependent variable, a one-way analysis of variance with repeated measures was conducted to find significant differences and Bonferroni post-hoc test was performed. Results: The results of this study showed that exercise intensity was reduced statistically as increased surface level on the upper body. Muscle activity of the upper rectus abdominis and biceps femoris for 30% of surface level was significantly higher than the corresponding values for flat surface. However, the opposite was found in the rectus femoris. In general, muscle activity of the lower rectus abdominis, erector spinae, external oblique abdominis, and gluteus maximus increased when surface level increased, but the differences were not significant. Conclusion: As a result, the increase in surface level of the body would change muscle activity of the upper body, indicating that different surface level of the upper body may cause significant effect on particular muscles to be more active during mountain climbing exercise. Based on results of this study, it is suggested to set up an appropriate surface level to target particular muscle to expect an effective training. It is also important to set adequate surface levels to create an effective training condition for preventing exercise injuries.

• Journal of the Korean Mathematical Society
• /
• v.50 no.5
• /
• pp.973-989
• /
• 2013

We observe from known results that the set of nilpotent elements in Armendariz rings has an important role. The upper nilradical coincides with the prime radical in Armendariz rings. So it can be shown that the factor ring of an Armendariz ring over its prime radical is also Armendariz, with the help of Antoine's results for nil-Armendariz rings. We study the structure of rings with such property in Armendariz rings and introduce APR as a generalization. It is shown that APR is placed between Armendariz and nil-Armendariz. It is shown that an APR ring which is not Armendariz, can always be constructed from any Armendariz ring. It is also proved that a ring R is APR if and only if so is R[$x$], and that N(R[$x$]) = N(R)[$x$] when R is APR, where R[$x$] is the polynomial ring with an indeterminate $x$ over R and N(-) denotes the set of all nilpotent elements. Several kinds of APR rings are found or constructed in the precess related to ordinary ring constructions.

• Journal of Fashion Business
• /
• v.10 no.1
• /
• pp.58-78
• /
• 2006

This study materializes body type measurement based on ISO(International Standard Organization)size system, targeting 1381 Chinese women, and suggests clothes size standards. It provides basic information on Chinese women size for Korean clothes companies advanced in China to strengthen the national competitiveness of Korean clothing industry. First, after the body type examination, based on the ISO's drop value, the results of types are in its decreasing order: A, M, H, and D. Second, women upper garment sizes by body types are set in the section of height 152cm, 160cm, and 168cm. Garment sizes are suggested based on the section that has more than 0.8% of appearance rate for the survey target: A, H, and M. Third, the clothes sizes of women pants are set in the section of height 152cm, 160cm, and 168cm just with the same as the upper garment.