• Archives of design research
• /
• v.17 no.1
• /
• pp.277-288
• /
• 2004

The purpose of this study is to find out the characteristics of building materials and exterior colors of the traditional palaces in Korea and Japan. Nijo-Castle in Japan and Changdeok-Palace in Korea were selected for the subjects of the study. For the color measurement and analysis, NCS color system was used. The results of this study are as follows. In Changdeok-Palace, Korean pine woods, granites, tiles, blue tiles, clays, Jeon-dol(Korean unique tiles for fences and ground), lime powders, straws, Korean traditional papers, bronze were used for building materials. In Nijo-Castle, however, Japanese pine woods, granites, tiles, clays, lime powders, straws, Japanese traditional papers, bronze and golds were used. As for hues in Changdeok-Palace, Y to R, G ∼ G30Y, R80B∼B, B to G were used mainly, and in Nijo-castle, Y to R, B80G∼B90G, G30Y were found. As for tones in Changdeok-Palace, every kinds of tones were used except 'Deep chromatic' and in Nijo-Castle, 'Greyish chromatic', 'Toned light grey', 'Dark deep', 'Toned dark grey', 'Toned grey' were used. At this study, the building materials and exterior colors were analyized between Changdeok-Palace and Nijo-Castle. And from the results, it is expected that we would understand different cultures of two nations, and get the concept of making their own unique characteristics.

• Journal of the Korean Society of Food Culture
• /
• v.23 no.4
• /
• pp.469-478
• /
• 2008

In this study, degree of rancidity and trans fatty acid formation were examined in fat and oils, including soybean oil (SB), canola oil (CA), corn germ oil (CO), olive oil (OL), palm oil (PO), and beef tallow (BT), during heating for 10-130 minutes at 160-200$^{\circ}C$. In order to determine the rancidity of the fat and oils, acid values (AV), iodine values (IV), viscosity, and color were measured. Changes in the amounts of fatty acids and the formation of trans fatty acids were measured using GC and HPLC. For all groups, AV increased, IV decreased, and coefficients of viscosity and color increased as the heating temperature and heating time increased, indicating there were positive correlations between the heating temperature and time and AV. In addition, all groups had similar amounts of trans fatty acids, with the exception of the beef tallow; however, its level only slightly increased with heating. The olive oil had the lowest trans fatty acid content and the lowest amount created by heating. The order of trans fatty acid amounts generated while heating was BT>PO>CO>CA>SB>OL. According to the study results, the deep frying temperature during cooking should be 160-180$^{\circ}C$ in order to reduce AV and the amount of trans fatty acids that are formed. In addition, it is better to remove beef tallow during cooking and avoid heating at high temperatures since it results in high levels of trans fatty acids. The correlation between the amount of trans fatty acids and AV was positive, while the correlation between the amount of trans fatty acids and IV was negative, indicating that AV and trans fatty acid levels increase while IV decreases as the deep frying temperature and time increase. From the results, it was found that reducing the deep frying temperature and time can lessen increases in AV and trans fatty acids, and decrease IV. Accordingly, to reduce AV and trans fatty acid formation, the ideal deep frying conditions would be to use olive oil or soybean oil rather than beef tallow or palm oil at a temperature of 160-180$^{\circ}C$.

• Korean Journal of Breeding Science
• /
• v.41 no.1
• /
• pp.1-8
• /
• 2009

This study was carried out to breed and develop high quality and functional nutrient tomato with multi disease resistance as well as a stable growing adaptation for fresh market usage under protected plastic houses cultivation. The materials were used 5 inbred lines and their 6 hybrids of large tomato group, which have been bred and developed from 1999 to 2007 in Division of Plant Resource Department of Chungnam National University. Fruit weight showed hybrid vigor effect that $F_1$ hybrids weighed more than their parent lines, fruit shape formed three type of oblate, deep oblate and globe shape, in firmness and pericarp thickness have got a high significant correlation, inbred DN611 line was measured the most firm fruit with 6.04 mm pericarp thickness. In fruit color at maturity, pink color crossed to red color appeared all red fruit color in the $F_1$ hybrids, it means red skin color is a dominant gene compared to pink skin color is a recessive gene in tomato, while between fruit skin color and shoulder part color showed no any co-relationship. The sugar content and titratable acid of $F_1$ hybrids inherited an intermediate data of their parent lines, the flavor of KP543 inbred line and the hybrid (JB535 x KP543) revealed the better taste with high brix and proper titratable acid content$^{*}$. In beta carotene content DN611 line showed 2~3 times higher than other materials so that its 3 hybrids contained an increased level of beta carotene, lycopene content was not so much difference among inbred lines and $F_1$ hybrids, of them MD508 contained higher of 8.72 mg and hybrid (JB535 x JA517) had 8.05 mg lycopene content per 100 g fruit, overall pink skin color and red skin color measured a higher lycopene content than yellow and orange skin color at ripe stage. In disease resistance test by PCR marker for Fusarium race2 (I2), Nematode (Mi1), ToMV ($Tm2^2$), Cladosporium (Cf9), (JB535 x JA517) hybrid have got multi-resistance with homozygote band in Nematode, ToMV, Cladosporium and heterozygote band in Fusarium race2. Through this breeding program we could select high quality and functional nutrient with multi resistant $F_1$ hybrids and inbred lines in tomato which are two best hybrids (JB535 x MD508), (JB535 x JA517), additionally developed high beta carotene inbred line DN611 and increased the level of lycopene inbred line MD508. These results will be very useful to make a high quality tomato variety continuously.

• Textile Coloration and Finishing
• /
• v.22 no.4
• /
• pp.341-348
• /
• 2010

Sodium sulfate is commonly added in reactive dyebath in order to increase substantivity of the reactive dye to cellulose fiber by reducing repulsion between anionic dye and fiber. While sodium sulfate is mostly used in solid form, it is inconvenient to dissolve a large amount of powder sodium sulfate. Furthermore, if there is undissolved salt in dyebath it might cause unlevel dyeing. In this study, sodium sulfate in liquid or solid form was used in dyeing of cellulose fabric with reactive dyes of three primary color and the effect of type or amount of sodium sulfate on dyeing and fastness properties was investigated. When the amount of sodium sulfate rose to 30-50 g/l, K/S value of the dyed fabric markedly increased; further rise in sodium sulfate concentration resulted in slow increase in K/S value. For light color, optimum amount was about 30 g/l in solid form and 50-100 g/l in liquid form while, for medium to deep color, it was 50 g/l and 100-150 g/l in solid and liquid form, respectively. When using each optimum amount of salt in solid or liquid form for medium color, shape of dyeing curve as well as exhaustion was similar to each other. On the whole, similar color fastness results were obtained regardless of type or amount of sodium sulfate.

• Fashion & Textile Research Journal
• /
• v.14 no.3
• /
• pp.492-500
• /
• 2012

Cocoa is a popular drink for children and contains healthy polyphenols however; a deep brown stain is left when cocoa is spilled over clothes. The main pigments in cocoa are anthocyanins that change in washing solutions with different alkalinity and metals. The removal and color changes in a cocoa stain after washing with various pH solutions and water hardness were studied. Alkalinity and the water hardness of washing solutions were important factors for the removal of cocoa stains. The removal of cocoa things in washing solutions without detergent was low (and even became negative after removal and darker) in solutions with a pH 9 and above. The cocoa stain was not removed and only the fabric color faded, although the cocoa stained cloth was washed with Korea tap water that has a pH of 7. The cocoa stain removal in detergent solutions was conspicuously higher than for only water. Even in detergent solutions, the cocoa stain removal decreased as water hardness increased. Cocoa stain removal was more effective and the color dimmest when the stained cloth was washed in a solution without the metal cations, and the bleach added with the detergent at a temperature of $40^{\circ}C$ and for longer than 20 minutes. Effective and economical equipment for tap water softening for a washing machine should be developed and used to improve cocoa stain removal.

• FLOWER RESEARCH JOURNAL
• /
• v.16 no.4
• /
• pp.284-290
• /
• 2008

In order to elucidate growth characteristics, physiological responses and leaf color changes of leaf-variegated Arundinaria munsuensis and Carex ciliato-marginata for variegata, These experiments were performed under four different light ragimes control(full sun), 40%, 70% and 85%. Plant height and leaf area became promoted as shading level increases in leaf-variegated Arundinaria munsuensis. Photosynthetic effect was the highest in 85% shading of the full sun. Thus, this plant could be growing in the deep shade condition. Plant growth and the leaf color changes were most obviously shown in the 40% shading level. In the Carex ciliatomarginata for. variegata growth status was the best and green or strong greenish yellow leaf color turned out to be much clearer in the 40% shading treatment. And photosynthetic activity was enhanced as the light intensity decreases.

• Journal of Korea Multimedia Society
• /
• v.23 no.4
• /
• pp.539-548
• /
• 2020

Deep learning technology is currently being used and applied in many different fields. Convolution neural network (CNN) is a method of artificial neural networks in deep learning, which is commonly used for analyzing different types of images through classification. In the conventional classification of histopathology images of prostate carcinomas, the rating of cancer is classified by human subjective observation. However, this approach has produced to some misdiagnosing of cancer grading. To solve this problem, CNN based classification method is proposed in this paper, to train the histological images and classify the prostate cancer grading into two classes of the benign and malignant. The CNN architecture used in this paper is based on the VGG models, which is specialized for image classification. However, color normalization was performed based on the contrast enhancement technique, and the normalized images were used for CNN training, to compare the classification results of both original and normalized images. In all cases, accuracy was over 90%, accuracy of the original was 96%, accuracy of other cases was higher, and loss was the lowest with 9%.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.61 no.3
• /
• pp.55-65
• /
• 2019

Although crack on concrete exists from its early formation, crack requires attention as it affects stiffness of structure and can lead demolition of structure as it grows. Detecting cracks on concrete is needed to take action prior to performance degradation of structure, and deep learning can be utilized for it. In this study, transfer learning, one of the deep learning techniques, was used to detect the crack, as the amount of crack's image data was limited. Pre-trained Inception-v3 was applied as a base model for the transfer learning. Web scrapping was utilized to fetch images of concrete wall with or without crack from web. In the recognition of crack, image post-process including changing size or removing color were applied. In the visualization of crack, source images divided into 30px, 50px or 100px size were used as input data, and different numbers of input data per category were applied for each case. With the results of visualized crack image, false positive and false negative errors were examined. Highest accuracy for the recognizing crack was achieved when the source images were adjusted into 224px size under gray-scale. In visualization, the result using 50 data per category under 100px interval size showed the smallest error. With regard to the false positive error, the best result was obtained using 400 data per category, and regarding to the false negative error, the case using 50 data per category showed the best result.

• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.29-35
• /
• 2021

License plate recognition plays a key role in intelligent transportation systems. Therefore, it is a very important process to efficiently detect the number and character areas. In this paper, we propose a method to effectively detect license plate number area by applying deep learning and semantic image segmentation algorithm. The proposed method is an algorithm that detects number and text areas directly from the license plate without preprocessing such as pixel projection. The license plate image was acquired from a fixed camera installed on the road, and was used in various real situations taking into account both weather and lighting changes. The input images was normalized to reduce the color change, and the deep learning neural networks used in the experiment were Vgg16, Vgg19, ResNet18, and ResNet50. To examine the performance of the proposed method, we experimented with 500 license plate images. 300 sheets were used for learning and 200 sheets were used for testing. As a result of computer simulation, it was the best when using ResNet50, and 95.77% accuracy was obtained.

• The Journal of Information Systems
• /
• v.31 no.4
• /
• pp.145-160
• /
• 2022

Purpose The purpose of this study is to develop a virtual try-on deep learning model that can efficiently learn front and back clothes images. It is expected that the application of virtual try-on clothing service in the fashion and textile industry field will be vitalization. Design/methodology/approach The data used in this study used 232,355 clothes and product images. The image data input to the model is divided into 5 categories: original clothing image and wearer image, clothing segmentation, wearer's body Densepose heatmap, wearer's clothing-agnosting. We advanced the HR-VITON model in the way of Mixed-Precison, Gradient Accumulation, and sharing model weights. Findings As a result of this study, we demonstrated that the weight-shared MP-GA HR-VITON model can efficiently learn front and back fashion images. As a result, this proposed model quantitatively improves the quality of the generated image compared to the existing technique, and natural fitting is possible in both front and back images. SSIM was 0.8385 and 0.9204 in CP-VTON and the proposed model, LPIPS 0.2133 and 0.0642, FID 74.5421 and 11.8463, and KID 0.064 and 0.006. Using the deep learning model of this study, it is possible to naturally fit one color clothes, but when there are complex pictures and logos as shown in <Figure 6>, an unnatural pattern occurred in the generated image. If it is advanced based on the transformer, this problem may also be improved.