• IEMEK Journal of Embedded Systems and Applications
• /
• 제10권5호
• /
• pp.325-333
• /
• 2015

Vision based night-time vehicle detection has been an emerging research field in various advanced driver assistance systems(ADAS) and automotive vehicle as well as automatic head-lamp control. In this paper, we propose night-time vehicle detection method based on multi-class support vector machine(SVM) that consists of thresholding, labeling, feature extraction, and multi-class SVM. Vehicle light candidate blobs are extracted by local mean based thresholding following by labeling process. Seven geometric and stochastic features are extracted from each candidate through the feature extraction step. Each candidate blob is classified into vehicle light or not by multi-class SVM. Four different multi-class SVM including one-against-all(OAA), one-against-one(OAO), top-down tree structured and bottom-up tree structured SVM classifiers are implemented and evaluated in terms of vehicle detection performances. Through the simulations tested on road video sequences, we prove that top-down tree structured and bottom-up tree structured SVM have relatively better performances than the others.

• Journal of the Korea Furniture Society
• /
• 제23권4호
• /
• pp.407-414
• /
• 2012

In order to develop the end use of cylindrical laminated veneer lumber (LVL) such as wooden crafts, the light resistance properties of natural dye-colored and finished cylindrical LVL was investigated. The study results were as follows; The cross section of LVL showed the higher light resistance than its tangential section, and the light resistance also increased with surface coated. On the other hand, LVL specimen were colored by 9 natural dyes of amur cork-tree etc., and the natural and apparent hue harmonized with wood was found after coating and light resistance test. Particularly, The dyes extracted from amur cork-tree, persimon and sappan wood showed orange color, reddish brown and red color respectively, giving an accenting effect to original wood color.

• ETRI Journal
• /
• 제28권4호
• /
• pp.517-520
• /
• 2006

In this letter, a new sharing mechanism, SRLG sharing, is proposed, which allows the links of the same shared risk link group (SRLG) in a primary light tree to share protections in WDM optical networks. In previous studies, how to share spare resources with SRLG constraints has not been studied in multicast optical networks. In this letter, considering SRLG sharing, we propose a novel algorithm -multicast with SRLG sharing (MSS)- to establish a protection light tree. Finally, the algorithm MSS and the algorithm multicast with no SRLG sharing (MNSS) are compared through a simulation to show that our new sharing scheme of SRLG sharing is more efficient than that of no SRLG sharing in terms of spare resource utilization and blocking probability.

• Journal of the Korean Society of Clothing and Textiles
• /
• 제29권6호
• /
• pp.868-876
• /
• 2005

Silk fabrics were dyed with yellow natural materials. Gardenia, turmeric, amur cork tree, safflower, Japanese pagoda tree, and onion were used as natural materials. Color image of natural dyed silk fabrics was classified by 4 factors(Cheerfulness, Comfortness, Pastrol, Revealation). Cheerfulness factor affected significantly with color image. Amur cork tree or turmeric dyed fabrics were shown as light and cute, safflower or gardenia dyed fabric shown as comfort, and onion dyed fabrics shown as mature and simple. Color image with specialty was significant difference in Preference and Revelation factor. Preference was appeared as amur cork tree>turmeric>gardenia>safflower>Japanese pagoda tree>onion.

• KIPS Transactions on Computer and Communication Systems
• /
• 제11권12호
• /
• pp.445-452
• /
• 2022

Class distribution of unbalanced data is an important part of the digital world and is a significant part of cybersecurity. Abnormal activity of unbalanced data should be found and problems solved. Although a system capable of tracking patterns in all transactions is needed, machine learning with disproportionate data, which typically has abnormal patterns, can ignore and degrade performance for minority layers, and predictive models can be inaccurately biased. In this paper, we predict target variables and improve accuracy by combining estimates using Synthetic Minority Oversampling Technique (SMOTE) and Light GBM algorithms as an approach to address unbalanced datasets. Experimental results were compared with logistic regression, decision tree, KNN, Random Forest, and XGBoost algorithms. The performance was similar in accuracy and reproduction rate, but in precision, two algorithms performed at Random Forest 80.76% and Light GBM 97.16%, and in F1-score, Random Forest 84.67% and Light GBM 91.96%. As a result of this experiment, it was confirmed that Light GBM's performance was similar without deviation or improved by up to 16% compared to five algorithms.

• Textile Coloration and Finishing
• /
• 제12권3호
• /
• pp.183-191
• /
• 2000

The dyeing of silk fabric with chinese scholar tree extract was investigated. The colorant was extracted with distilled water, and concentrated under reduced pressure, and, finally spray-dried. The optimum temperature and dyeing time for the dyeing of silk with the scholar tree extract was $60^\circ{C}$ and 60 minutes respectively. In mordant dyeing, pre-mordanting method was most effective for color yield, and 1~2 % of mordant concentration was acceptable. Fastness of the dyed silk fabric to laundering and rubbing was shown to be good, but fastness to light, dry cleaning, and perspiration were varied with used mordants.

• Fashion & Textile Research Journal
• /
• 제10권4호
• /
• pp.544-551
• /
• 2008

In this study, the colorants of Amur cork tree were extracted with boiling water. Chitosan crosslinked cotton fabrics have been dyed with aqueous extract of Amur cork tree and their dyeabilities on the fabrics were studied. Additionally the fastness to washing and light were also investigated. Cotton fabrics were treated with a crosslinking agent epichlorohydrin in the presence of chitosan to provide the cotton fabrics the dyeing properties of natural dye(Amur cork tree) by the chemical linking of chitosan to the cellulose structure. This process was applied by means of the conventional mercerizing process. The chitosan finishing and durable press finishing of the cotton fabrics occurred simultaneously in the mercerization bath. On the surface color change, the fabric of no-chitosan finished and no-mordanted has greenish yellow. The more crosslinked chitosan on cotton fabrics has the more turned down greenish on the surface color, as increasing the concentration of chitosan, greenish color turn down to the yellow close the 90o hue angle. In all sorts of fabrics, dyeability(K/S) is slightly affected by the number of manufacturing process and the concentration of chitosan. But only mercerized cotton fabric has higher dyeability (K/S) than mordant treated cotton fabrics. Wash fastness has little different results by each condition, but almost similar values. Light fastness was improved with chitosan treatment on cotton fabric.

• Journal of Fashion Business
• /
• 제10권2호
• /
• pp.27-39
• /
• 2006

In this study, physical/chemical characteristics of the fabrics dyed using Japanese pagoda tree which is a yellowish dyestuff was investigated. We tried to estimate the dyeing mechanism among fiber macromolecules, mordants, chitosan, and dyestuffs by measuring the change of air-permeability according to the dyeing process. Also wash fastness and light fastness were measured in order to estimate the fastness according to the dyeing characteristics. When the chitosan pre-treatment was introduced, the wash fastness increased significantly in the cotton fabrics while the fastness did not change in the nylon fabrics significantly. The light fastness was not improved by the chitosan treatment. when the antibacterial activity was measured for the dyed cotton fabrics, Japanese pagoda tree itself did not exhibit antibacterial activity. However, chitosan treated dyed fabrics exhibited low antimicrobial activity.

• Journal of the Korean Society of Safety
• /
• 제24권4호
• /
• pp.1-10
• /
• 2009

The characteristics of smoke emissions from diesel heavy duty vehicles which weigh over 5.5 tons was investigated by driving the vehicles with both the Lugdown 3 modes in the chassis dynamometer and tree accelerating mode under no load. The vehicles include commercial vehicles such as bus, microbus, trucks and specialized vehicles, etc. The total numbers of the vehicles tested were 200. The light extinction method was used to measure the smoke emissions from the vehicles tail pipe. The values of the smoke emissions in the tree accelerating mode showed $0{\sim}20%$ band nearly independent of both the mileage and year of production of the tested vehicles, while those in the Lugdown 3 modes showed $0{\sim}99%$ of wide band. The correlation coefficients between the values of the smoke emissions with both the Lugdown 3 modes and the free acceleration mode were 0.12, 0.08, 0.12, respectively. The inspection with Lugdown 3 modes is better one than that with tree acceleration from the point of exact inspection of the diesel vehicles' smoke emission.

• Textile Coloration and Finishing
• /
• 제11권1호
• /
• pp.9-15
• /
• 1999

Using water and methanol, the colorant components were extracted from the heartwood of amor cork tree which is known as the oldest natural yellow dye. The dyeabilities of the colorant on cotton and the fastness of cotton fabrics dyed with amur cork tree extracts were investigated. According to the dye uptake and mordant concentration, antibacterial activities and deodorization properties of dyed fabrics were also investigated. As a results, K/S values of dyed fabrics were increased with concentrations of mordants and amur cork tree extracts. Washing fastness of dyed fabrics was very low, light fastness was fair and the other fastness were good. Deodorization rates of dyed fabrics were increased in proportion to concentrations of amur cork tree extracts. The reduction rates of bacteria measured by Bioassay tests were also increased in proportion to concentrations of mordants and amur cork tree extracts.