• Journal of Korean Society for Atmospheric Environment
• /
• v.22 no.4
• /
• pp.440-450
• /
• 2006

PM10, NOx, and $O_3$ were measured at six locations, of which each three is horizontally and vertically distributed respectively, in an apartment complex around the heavily traffic road. Those were measured seven times a day with two hours interval starting from 8 o'clock in the morning for 15 days during May 2005 $\sim$ September 2005. PM10 and NOx showed high concentrations in rush hours while low concentrations in midday due to the direct emissions from automobiles in operation. Temporal variations of 01 showed very much similar trend appeared in normal urban atmospheres. The spatial distributions of PM10, NOx and $O_3$ showed that almost all of concentrations were higher in a row of Roadside > Surface at 130 m apart from the road > Surface at 230 m apart from the road > 3rd floor of apartment building > 15th floor of apartment building > 27th floor of apartment building. Model equations, which can project spatial concentration distributions, were constructed by combining the horizontal and the vertical linear regression equations derived from six mean values corresponding to six measuring locations. According to inter-comparison of PM10, NOx, and $O_3$ with the constructed model equations, concentration gradients were higher in a row of Vertical direction of NOx > Vertical direction of PM10 > Horizontal direction of NOx > Horizontal direction of PMIO > Vertical direction of $O_3$ > Horizontal direction of $O_3$. Why concentration gradient of particulate PM10 is lower than that of gaseous NOx is in question, and should be studied.

• Journal of Broadcast Engineering
• /
• v.15 no.5
• /
• pp.697-702
• /
• 2010

Contour artifact is known as the unintentional result of quantizing a flat area that has smooth gradients. In this letter, a decontouring algorithm is proposed to efficiently remove false contours that occur in typical block-based video coding applications. First, the algorithm goes through a refinement stage to determine candidate blocks probably having noticeable false contours with different kinds of features in a block. Then, pseudo-random noise masking is applied to those blocks to mitigate the contour artifacts. This block-based selective decontouring can efficiently remove the unnecessary processing of those blocks that have no false contour, which incidentally ensures a minor penalty in visual quality and computational complexity. The proposed algorithm was demonstrated, integrated into H.264/AVC, that visual quality can be significantly enhanced with an ignorable rate-distortion (RD) loss and an minor increase in computational complexity.

• BMB Reports
• /
• v.43 no.3
• /
• pp.182-187
• /
• 2010

family (CMTM) is a novel family of proteins linking classical chemokines and the transmembrane 4 superfamily (TM4SF). Our earlier studies indicated several CMTM members (such as CKLF1 and CMTM2) have a secreted form. This is the first report of the secreted form of CMTM5-v1, the major RNA splicing form of CMTM5, which is produced as small vesicles (<100 nm diameter) and floats at a peak density of 1.19 g/ml on continuous sucrose gradients. CMTM5-v1 has no obvious co-localization with CD63 or Golgi complex. In addition, brefeldin A but not wortmannin can inhibit the secretion of CMTM5-v1. Our results suggest that CMTM5-v1 might be secreted via a different vesicle-mediated secretory pathway, which will be helpful for the studies of vesicle-mediated secretion and MARVEL domain-containing proteins.

• Korean Journal of Materials Research
• /
• v.12 no.12
• /
• pp.897-903
• /
• 2002

Morphological evolution and growth mechanism at the solid/liquid interface during solidification were investigated in the Ni-base superalloy GTD111M by directional soldification and quenching(DSQ) technique. The experiments were conducted by changing solidification rate(V) and thermal gradient(G) which are major solidification process variables. High thermal gradient condition could be obtained by increasing the furnace temperature and closely attaching the heating and cooling zones in the Bridgeman type furnace. The dendritic/equiaxed transition was found in the G/V value lower than $0.05$\times$10{^3}^{\circ}C$s/$\textrm{mm}^2$, and the planar interface of the MC-${\gamma}$ eutectic was found under $17 $\times$ 10{^3}^{\circ}C$ s/$\textrm{mm}^2$. It was confirmed that the dendrite spacing depended on the cooling rate(GV), and the primary spacing was affected by the thermal gradient more than solidification rate. The dendrite lengths were decreased as increasing the thermal graditne, and the dendrite tip temperature was close to the liquidus temperature at $50 \mu\textrm{m}$/s.

• The KIPS Transactions:PartB
• /
• v.13B no.3 s.106
• /
• pp.197-204
• /
• 2006

Recent advances in information technology have resulted in rapid growth in the mobile communication. With this explosive growth, reliable transmission and error concealment technique become increasingly important to offer high quality multimedia services. In this paper, we propose an improved BMA(Boundary Matching Algorithm) method using gradient vectors to conceal channel errors in inter-frames of H.264 video images. General BMA method computes the sum of pixel differences of adjacent pixels of the candidate block and its neighbouring blocks, assuming that adjacent pixels have almost the same value. In real images, however, there exist some gradients, which means that the pixel values are increasing or decreasing in a specific direction. In this paper, we develop a precise estimation method of errors in candidates blocks using gradient information and try to recover lost blocks with this technique. Experimental results show the improvement of picture quality about $1{\sim}3dB$ compared to existing methods.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.11
• /
• pp.680-688
• /
• 2005

The electrical current generated by heart creates not only electric potential but also a magnetic field. We have observed electrophysiological phenomena of the heart by measuring components of magnetocardiogram(MCG) using 61 channel superconducting quantum interference device(SQUD) system. We have analyzed the possibility and characteristics of MCG parameters for diagnosis of ischemic heart disease. A technique for automatic analysis of MCG signals in time domain was developed. The methods for detecting the position, the interval, the amplitude ratio, and the direction of single current dipole were examined in the MCG wave. The position and interval parameters were obtained by calculating the gradients of a envelope curve which could be formed by the difference between the maximum and minimum envelope of multi-channel MCG signals. We show some differences of the frequency contour map between the normal MCG and the abnormal (ischemic heart disease) MCG. The direction of single current dipole can be defined by rotating the magnetic field according to Biot-Savart's law at each point of MCG signals. In this study, we have examined the direction of single current dipole from searching for the centroids of positive and negative magnetic fields. The amplitude ratio parameters for measuring 57 deviation consisted of A$_{T}$/A$_{R}$ and other ratios. and We developed a new analysis method, which is based on the frequency contour map of electromagnetic field. Using theses parameters, we founded significant differences between normal subjects and ischemic patients in some parameters.

• Korean Journal of Cognitive Science
• /
• v.10 no.3
• /
• pp.1-15
• /
• 1999

Based on known physiological and psychophysical results, a neural network model for visual selection, called FeaureGate is proposed. The model consists of a hierarchy of spatial maps. and the flow of information from each level of the hierarchy to the next is controlled by attentional gates. The gates are jointly controlled by a bottom-up system favoring locations with unique features. and a top-down mechanism favoring locations with features designated as target features. The present study focuses on the top-down mechanism of the FeatureGate model that produces results similar to Moran and Desimone's (1985), which many current models have failed to explain, The FeatureGate model allows a consistent interpretation of many different experimental results in visual attention. including parallel feature searches and serial conjunction searches. attentional gradients triggered by cuing, feature-driven spatial selection, split a attention, inhibition of distractor locations, and flanking inhibition. This framework can be extended to produce a model of shape recognition using upper-level units that respond to configurations of features.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.893-898
• /
• 2005

Due to the merits that only a small amount of computation is needed for solutions and stochastic policies can be handled explicitly, the actor-critic algorithm, which is a class of reinforcement learning methods, has recently attracted a lot of interests in the area of artificial intelligence. The actor-critic network composes of tile actor network for selecting control inputs and the critic network for estimating value functions, and in its training stage, the actor and critic networks take the strategy, of changing their parameters adaptively in order to select excellent control inputs and yield accurate approximation for value functions as fast as possible. In this paper, we consider a new actor-critic algorithm employing an RLS(Recursive Least Square) method for critic learning, and policy gradients for actor learning. The applicability of the considered algorithm is illustrated with experiments on the two linked robot arm.

• Journal of Korean Society of Forest Science
• /
• v.90 no.3
• /
• pp.324-330
• /
• 2001

This study was carried out to develop prediction model for fill slope failure of forest road in igneous rock area using fuzzy theory which is non-linear model. The results were summarized as follows. The importance weight of factors on fill slope failure was ranked in the order of fill slope length, fill slope gradient, soil type, aspect, road position and longitudinal slope form. The degree of potential slope failure was high mainly under the such conditions as fill slope length greater than 8m, fill slope gradients steeper than $40^{\circ}$, constituent material with weathered rock, aspect of NE and road on ridge position. The optimal prediction model was developed with 0.15 of optimal coefficient(c) and 3.1165 of ${\lambda}$-value when fuzzy integral value of slope failure possibility is more than 0.5. And the discriminant accuracy was 86.8%, which shows the high availability for discrimination.

• Composites Research
• /
• v.31 no.1
• /
• pp.12-16
• /
• 2018

Composite material and mold steel can be expanded differently with the temperature gradients during the forming process because their coefficients of thermal expansions are not the same. Therefore, in order to manufacture the product with accuracy, it is necessary to verify that the forming pressure on the surface of the composite material is maintained to the required level from the material supplier. In this paper, the pressure between the composite material and mold due to the temperature difference was predicted by finite element analysis and the accuracy of predicted value was verified by measuring the thermal expansions of mold steel by the ruler. The predicted value by finite element analysis is closely in agreement with one by the experiment within the required tolerance value of ${\pm}0.05mm$.