• Journal of Korea Multimedia Society
• /
• v.13 no.11
• /
• pp.1698-1705
• /
• 2010

In this paper, we propose a fast liver segmentation method from magnetic resonance(MR) images. Our method efficiently divides a MR image into a set of discrete objects, and boundaries based on the normalized gradient magnitude information. Then, the objects belonging to the liver are detected by using 2D seeded region growing with seed points, which are extracted from the segmented liver region of the slice immediately above or below the current slice. Finally, rolling ball algorithm, and connected component analysis minimizes false positive error near the liver boundaries. Our method was validated by twenty data sets and the results were compared with the manually segmented result. The average volumetric overlap error was 5.2%, and average absolute volumetric measurement error was 1.9%. The average processing time for segmenting one data set was about three seconds. Our method could be used for computer-aided liver diagnosis, which requires a fast and accurate segmentation of liver.

• ALGAE
• /
• v.35 no.4
• /
• pp.361-373
• /
• 2020

The community dynamics of benthic diatoms in the hypersaline environment are investigated to advance our understanding how salinity impacts marine life. Diatoms were sampled in the two salterns encompassing salt Ponds, ditches, and seawater reservoirs (n = 11), along the salinity gradient (max = 324 psu), and nearby tidal flats (n = 2). The floral assemblages and distributions across sites and stations showed great variations, with a total of 169 identified taxa. First, not surprisingly, higher diversity of benthic diatoms was found at natural tidal flats than salterns. The saltern diatoms generally showed salinity dependent distributions with distinct spatial changes in species composition and dominant taxa. Biota-environment and principal component analysis confirmed that salinity, mud content, and total nitrogen were key factors influencing the overall benthic community structure. Some dominant species, e.g., Nitzschia scalpelliformis and Achnanthes sp. 1, showed salinity tolerance / preference. The number of diatom species at salinity of >100 psu reduced over half and no diatoms were found at maximum salinity of 324 psu. The highest salinity for the observed live diatoms was 205 psu, however, a simple regression indicated a theoretical salinity threshold of ~300 psu on the survival. Finally, the indicator species were identified along the salinity gradient in salterns as well as natural tidal flats. Overall, high species numbers, varying taxa, and euryhaline distributions of saltern diatoms collectively reflected a dynamic saltern ecosystem. The present study would provide backgrounds for biodiversity monitoring of ecologically important microalgal producers in some unique hypersaline environment, and elsewhere.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.104-123
• /
• 2022

Style transfer based on neural network provides very high quality results by reflecting the high level structural characteristics of images, and thereby has recently attracted great attention. This paper deals with the problem of resolution limitation due to GPU memory in performing such neural style transfer. We can expect that the gradient operation for style transfer based on partial image, with the aid of the fixed size of receptive field, can produce the same result as the gradient operation using the entire image. Based on this idea, each component of the style transfer loss function is analyzed in this paper to obtain the necessary conditions for partitioning and padding, and to identify, among the information required for gradient calculation, the one that depends on the entire input. By structuring such information for using it as auxiliary constant input for partition-based gradient calculation, this paper develops a recursive algorithm for super high-resolution image style transfer. Since the proposed method performs style transfer by partitioning input image into the size that a GPU can handle, it can perform style transfer without the limit of the input image resolution accompanied by the GPU memory size. With the aid of such super high-resolution support, the proposed method can provide a unique style characteristics of detailed area which can only be appreciated in super high-resolution style transfer.

• Applied Biological Chemistry
• /
• v.13 no.1
• /
• pp.59-64
• /
• 1970

Human Bence Jones Protein could be purified by DEAF-Sephadex A-50 column $(2{\times}37cm)$ with 0.02M phosphate Buffer (pH 8.0) and gradient increasing with NaCl concentration as in Fig. 2-4. Sample As (K-type Bence Jones Protein) had two component, F-I was major component and its dried weight was 350mg. of starting material of 500mg. Other Sample Im and Ik (${\lambda}$-type Bence Jones Protein) was purified by DEAE-Sephadex A-50 with 0.02M phosphate Buffer(pH 8.0)too. F-I (major component) of Im and F-I of Ik were 242mg and 146mg. its dried weight respectively. K-type of Bence Jones Protein's(As, Ko, Ta.) N-terminal amino acid residue was determined by method of DNP,. K-type of Bence Jones Protein's amino acid residue were either glutamic acid or aspartic acid. Sample Ta was confirmed as glutamic acid its N-Terminal. As and Ko were aspartic acid. Each yellowish spot (DNP-amino acids) were extracted with 4ml. of pH 8.05% $NaHCO_3$ solution and calculated its recovery by O.D. $(360m{\mu}$ using the ${\varepsilon}=18.1{\times}10^3DNP$ $Asp\;{\varepsilon}=17.41{\times}10^(3)\;DNP\;Glu$ considering 50% lose during; the acid (6N-HCI) hydrolysis. Recovery of ko and As were 54.3% and 65% of its starting materials (DNP-Protein). Sample Ta's recovery was 85% of its DNP-protein. ${\lambda}$-type of Bence Jones Protein was rot investigated its N-terminal amino acid residue by DNP-method, probably it was blocked its N-terminal residue with glutamic acid.

• Geophysics and Geophysical Exploration
• /
• v.14 no.3
• /
• pp.220-226
• /
• 2011

In the elastic wave equations, both horizontal and vertical displacements are defined. Since we can measure both the horizontal and vertical displacements in field acquisition, these displacements compose a displacement vector. In this study, we propose a frequency-domain elastic waveform inversion technique taking advantage of the magnitudes of displacement vectors to define objective function. When we apply this displacement-vector objective function to the frequency-domain waveform inversion, the inversion process naturally incorporates the back-propagation algorithm. Through the inversion examples with the Marmousi model and the SEG/EAGE salt model, we could note that the RMS error of the solution obtained by our algorithm decreased more stably than that of the conventional method. Particularly, the density of the Marmousi model and the low-velocity sub-salt zone of the SEG/EAGE salt model were successfully recovered. Since the gradient direction obtained from the proposed objective function is numerically unstable, we need additional study to stabilize the gradient direction. In order to perform the waveform inversion using the displacementvector objective function, it is necessary to acquire multi-component data. Hence, more rigorous study should be continued for the multi-component land acquisition or OBC (Ocean Bottom Cable) multi-component survey.

• KOREAN JOURNAL OF CROP SCIENCE
• /
• v.44 no.1
• /
• pp.78-85
• /
• 1999

To investigate kernel hardness, a compression test which is widely used to measure the hardness of individual kernels as a physical testing method was made simultaneously with the measurement of friabilin (15KDa) which is strongly associated with kernel hardness and was recently developed as a biochemical marker for evaluating kernel hardness in 79 Korean wheat varieties and experimental lines. With the scattered diagram based on the principal component analysis from the parameters of the compression test, 79 Korean wheat varieties were classified into three groups based on the principal component analysis. Since conventional methods required large amount of flour samples for analysis of friabilin due to the relatively small amount of friabilin in wheat kernels, those methods had limitations for quality prediction in wheat breeding programs. An extraction of friabilin from the starch of a single kernel through cesium chloride gradient centrifugation was successful in this experiment. Among 79 Korean wheat varieties and experimental lines 50 lines (63.3%) exhibited a friabilin band and 29 lines (36.7%) did not show a friabilin band. In this study, lines that contained high maximum force and the lower ratio of minimum force to maximum force showed the absence of the friabilin band. Identification of friabilin, which is the product of a major gene, could be applied in the screening procedures of kernel hardness. The single kernel analysis system for friabilin was found to be an easy, simple and effective screening method for early generation materials in a wheat breeding program for quality improvement.

• Archives of Pharmacal Research
• /
• v.21 no.5
• /
• pp.521-526
• /
• 1998

The physicochmical properties of recombinant hepatitis B surface antigen (r-HBsAg), which was expressed in C127 mammalian cell were studied. Using roller bottle culture in DMEM supplemented with fetal bovine serum, 10-15 mg/L of r-HBsAg was produced with about 31% of purification yield. The purity of r-HBsAg by HPLC was 99.8% and electron microscopic examination showed homogeneous spherical particle with 22 nm in diameter, a morphological characteristic of HBsAg. The density of r-HBsAg by CsCI density gradient method was 1.19g/ml and the isoelectric point by Mono $P^{TM}$ HR 5/20 column was 4.6. The analysis of subunit protein pattern using SDS-PAGE followed by scanning densitometry gave 81.3% of S protein and 18.7% of pre-S protein. fluorophore-assisted-carbohydrate-electrophoresis analysis showed the relative amount of carbohydrate to protein was 1.7% and it smajr component was N-acetyl glucosamine, which was about 39% of total carbohydrate. The relative amount of lipid to protein determined by vanillin phosphoric acid method was 32.5% and its major component was phospholipid, which was about 70% of total lipid. The physicochemical properties of C127 mammalian cell-derved r-HBsAg are similar to those of p-HBsAg, suggesting that the r-HBsAg can be used in developing a new preventive vaccine against hepatitis B.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.8
• /
• pp.307-324
• /
• 2018

Recently, recurrent neural networks have been attracting attention in solving prediction problem of sequential data through structure considering time dependency. However, as the time step of sequential data increases, the problem of the gradient vanishing is occurred. Long short-term memory models have been proposed to solve this problem, but there is a limit to storing a lot of data and preserving it for a long time. Therefore, research on memory-augmented neural network (MANN), which is a learning model using recurrent neural networks and memory elements, has been actively conducted. In this paper, we describe the structure and characteristics of MANN models that emerged as a hot topic in deep learning field and present the latest techniques and future research that utilize MANN.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.19 no.1
• /
• pp.39-47
• /
• 2009

The effects of thermo-solutal convection on mercurous chloride system of ${Hg_2}{Cl_2}$, and Ne during physical vapor transport are numerically investigated for further understanding and insight into essence of transport phenomena, For $10\;K{\le}{\Delta}T{\le}30\;K$, the growth rate slowly increases and, then is decreased gradually until ${\Delta}T$=50 K, The occurrence of this critical point near at ${\Delta}T$=30 K is likely to be due to the effects of thermo-physical properties stronger than the temperature gradient corresponding to driving force for thermal convection. For the range of $10\;Torr{\le}P_B{\le}300\;Torr$, the rate is second order-exponentially decayed with partial pressures of component B, $P_B$. For the range of $5{\le}M_B{\le}200$, the rate is second order-exponentially decayed with a function of molecular weight of component B, $M_B$. Like the case of a partial pressure of component B, the effects of a molecular weight arc: reflected through the binary diffusivity coefficients, which are intimately related with suppressing the convection flow inside the growth enclosure, i,e., transition from convection to diffusion-dominant flow mode as the molecular weight of B increases. The convective mode is near at a ground level, i,e., on earth (1 $g_0$), and the convection is switched to the diffusion mode for $0.1\;g_0{\le}g{\le}10^{-2}g_0$, whereas the diffusion region ranges from $10^{-2}g_0$ up to $10^{-5}g_0$.

• Applied Chemistry for Engineering
• /
• v.5 no.1
• /
• pp.127-138
• /
• 1994

In the process of pervaporation separation for aqueous ethanol solution through cellulose tai-acetate(CTA) membrane, the modelling on the solution-diffusion permeation mechanism was built up on the basis of sorption and permeation experimental results. Also its function type and parameter were examined. The composition of sorption equilibrium in three component system(Ethanol/Water/CTA) were compared with the calculated value by Flory-Huggins' equation using the pure component sorption data. In order to apply the thermodynamic equilibrium relationship between the membrane free composition in the membrane and the equilibrium composition in the liquid phase, the apparent activity this system, however, the results were not satisfied. Diffusion equations were expressed with the concentration gradient considering permeate alone, and a concentration-dependent diffusion coefficient which includes a parameter was used. And this model was fitted with the measured permeation rates. If the permeation rate and the amount of sorption of one component were much larger than those of the other, the bulk flow term could not be negligible. The flux and selectivity were increased with increasing temperature, and with decreasing downstream pressure.