• Journal of IKEEE
• /
• v.25 no.2
• /
• pp.273-279
• /
• 2021

In this paper, we propose an algorithm that shortens the learning time by performing individual learning using partitioning the deep learning structure. The proposed algorithm consists of four processes: network classification origin setting process, feature vector extraction process, feature noise removal process, and class classification process. First, in the process of setting the network classification starting point, the division starting point of the network structure for effective feature vector extraction is set. Second, in the feature vector extraction process, feature vectors are extracted without additional learning using the weights previously learned. Third, in the feature noise removal process, the extracted feature vector is received and the output value of each class is learned to remove noise from the data. Fourth, in the class classification process, the noise-removed feature vector is input to the multi-layer perceptron structure, and the result is output and learned. To evaluate the performance of the proposed algorithm, we experimented with the Extended Yale B face database. As a result of the experiment, in the case of the time required for one-time learning, the proposed algorithm reduced 40.7% based on the existing algorithm. In addition, the number of learning up to the target recognition rate was shortened compared with the existing algorithm. Through the experimental results, it was confirmed that the one-time learning time and the total learning time were reduced and improved over the existing algorithm.

• Journal of Bio-Environment Control
• /
• v.24 no.3
• /
• pp.243-251
• /
• 2015

This study aimed to determine the effects of root zone cooling using air duct on air temperature distribution and root zone and leaf temperatures of sweet pepper (Capsicum annum L. 'Veyron') grown on coir substrate hydroponic system in a greenhouse. When the air duct was laid at the passage adjacent the slab, the direction of air blowing was upstream at $45^{\circ}$. The cooling temperature was set at $20^{\circ}C$ for day and $18^{\circ}C$ for night. For cooing timing treatments, the cooling air was applied at all day (All-day), only night time (5 p.m. to 1 a.m.; Night), or no cooling (Control). The air temperature inside the greenhouse at a height of 40 and 80cm above the floor, and substrate and leaf temperatures, fruit characteristics, and fruit ratio were measured. Under the All-day treatment, the air temperature was decreased about $4.4{\sim}5.1^{\circ}C$ at the height of 40cm and $2.1{\sim}3.1^{\circ}C$ at the height of 80cm. Under the Night treatment, the air temperature was decreased about $3.4{\sim}3.8^{\circ}C$ at the height of 40cm and $2.2{\sim}2.7^{\circ}C$ at the height of 80cm. The daily average temperature in the substrate was in the order of the Control ($27.7^{\circ}C$) > Night ($24.1^{\circ}C$) > All-day ($22.8^{\circ}C$) treatment. Cooling the passage with either upstream blowing at $45^{\circ}$ or horizontal blowing at $180^{\circ}$ was effective in lowering the air temperature at a height of 50cm; however, no difference at a height of 100cm. Cooling the passage with perpendicular direction at $90^{\circ}$ was effective in lowering the air temperature at the height between 100 and 200cm above the floor; however, no effect on the temperature at the height of 50cm. A greater decrease in leaf temperature was found at 7 p.m. than that at 9. a.m. under both All-day and Night treatments. Fresh weight partitioning of fruit was in the order of the All-day (48.6%) > Night (45.6%) > Control (24.4%) treatment. A higher fruit production was observed under the All-day treatment, in which the accumulated average temperature was the lowest, and it may have been led to a higher proportion of photosynthate distributed to fruit than other treatments.

• Korean Journal of Food Science and Technology
• /
• v.44 no.3
• /
• pp.263-268
• /
• 2012

Novobiocin is a coumarin-containing antibiotic, and has a longer half-life in various animals than other veterinary medicines. A simple and rapid high-performance liquid chromatography assay for the determination of residual novobiocin levels in chicken, beef and milk has been developed and validated. The separation condition for HPLC/UVD was optimized by a MG II $C_{18}$ (4.6 mm $ID{\times}250$ mm, 5 ${\mu}m$) column with 0.1% formic acid in $H_2O$/0.1% formic acid in Acetonitrile (40/60, v/v) as the mobile phase at a flow rate of 1.0 mL/min and the detection wavelength was set at 340 nm. Residues were extracted from tissue by blending with methanol. After liquid-liquid partitioning, lipid materials were removed with n-hexane and purification as Silica (1 g, 6 mL) cartridge with 10 mL acetone/dichloromethane (10/90, v/v). Limit of quantification and linearity performed by the analytical method were 0.02 mg/kg and 0.999 ($r^2$), and the recovery range was $88.8{\pm}5.6-100.3{\pm}4.4$, $88.8{\pm}7.2-97.0{\pm}3.2$ and $88.1{\pm}4.3-92.8{\pm}3.6%$. It is expected that this analytical method with regards to novobiocin in chicken, beef and milk could be applied as an official method to administer food safety on veterinary medicines.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.16 no.3
• /
• pp.219-226
• /
• 2014

Evaluation of tree responses to temperature elevation is critical for a development of forest management techniques coping with climate change. We conducted a study on the growth responses of Betula costata, Fraxinus rhynchophylla, and Quercus variabilis seedlings to open-field artificial warming. Artificial warming set-up using infra-red heater was built in 2012 and the temperature in warmed plots was regulated to be consistently $3^{\circ}C$ higher than that of control plots. The seeds of three species were sown, and the responses of growth, biomass allocation, and net photosynthetic rate of newly-germinated seedlings on the open-field artificial warming were determined. As a result, the growth responses of the seedlings differed with the species. B. costata showed decreases in the height to diameter ratio (H/D ratio), biomass, root weight to shoot weight ratio, and net photosynthetic rate. However, root collar diameter (RCD), height, biomass, and net photosynthetic rate of Q. variabilis were increased, while the response of F. rhynchophylla was rather obscure. There was no significant difference between warmed and control plots in seedling growth for 3 species in July, whereas, RCD, height, and H/D ratio of Q. variabilis were increased and H/D ratio of B. costata was decreased in November under warming. Species-specific growth responses to warming were similar to the species-specific responses of net photosynthetic rate and biomass allocation; therefore, net photosynthetic rate and biomass allocation might attribute to growth responses to warming. Besides, a relatively obvious response in autumn compared to summer might be affected by the phenological change following artificial warming. Species-specific responses of three deciduous species to warming in this study could be applied to the development of adaptive forest management policies to climate change.