• Journal of the Korea Society of Computer and Information
• /
• v.26 no.11
• /
• pp.173-182
• /
• 2021

In this paper, we analyzed driver's and passenger's motions that cause driver's distraction, and recognized 10 driver's behaviors related to mobile phones. First, distraction-inducing behaviors were classified into environments and factors, and related recent papers were analyzed. Based on the analyzed papers, 10 driver's behaviors related to cell phones, which are the main causes of distraction, were recognized. The experiment was conducted based on about 100,000 image data. Features were extracted through SURF and tested with three models (CNN, ResNet-101, and improved ResNet-101). The improved ResNet-101 model reduced training and validation errors by 8.2 times and 44.6 times compared to CNN, and the average precision and f1-score were maintained at a high level of 0.98. In addition, using CAM (class activation maps), it was reviewed whether the deep learning model used the cell phone object and location as the decisive cause when judging the driver's distraction behavior.

• Journal of Navigation and Port Research
• /
• v.47 no.6
• /
• pp.376-385
• /
• 2023

Fishing boat capsizing accidents account for more than half of all capsize accidents. These can occur for a variety of reasons, including inexperienced operation, bad weather, and poor maintenance. Due to the size and influence of the industry, technological complexity, and regional diversity, fishing ships are relatively under-researched compared to commercial ships. This study aimed to predict the rolling motion time series of fishing boats using an image-based deep learning model. Image-based deep learning can achieve high performance by learning various patterns in a time series. Three image-based deep learning models were used for this purpose: Xception, ResNet50, and CRNN. Xception and ResNet50 are composed of 177 and 184 layers, respectively, while CRNN is composed of 22 relatively thin layers. The experimental results showed that the Xception deep learning model recorded the lowest Symmetric mean absolute percentage error(sMAPE) of 0.04291 and Root Mean Squared Error(RMSE) of 0.0198. ResNet50 and CRNN recorded an RMSE of 0.0217 and 0.022, respectively. This confirms that the models with relatively deeper layers had higher accuracy.

• Journal of the Korean Magnetics Society
• /
• v.14 no.4
• /
• pp.143-148
• /
• 2004

Effects of residual flux density M$_{res}$ and number of inspection on the detection voltage and flux density B of the gas pipeline were investigated in MFL inspection, which is widely used for the non-destructive inspection in a gas pipeline. A simulation equipment composed of the magnetizer and iron ring attached on an aluminum disc was constructed instead of a huge gas pipeline facility. With this system. the iron ring could be perfectly demagnetized and signals from the bolt screw stuck on the disc could be clearly detected so that the effects of M$_{res}$S and the inspection number on the detection voltage and B of iron ring were effectively investigated. With increasing the number of inspection, M$_{res}$, B of the iron ring and the detection voltage decreased and then kept at constant values while final M$_{res}$ increased with increasing initial M$_{res}$. If inspection condition were kept unchanged, the detection voltage was proportional to the last M$_{res}$ of the iron ring instead of B. This was probably due to magnetic hysteresis of the iron ring inherited from magnetic domain so that consideration on the magnetic hysteresis was inevitable in the analysis of MFL signal from defects of a gas pipeline. A new inspection scheme using the magnetizer with reversed magnetization in the subsequent inspection was proposed from the result that a high detection voltage could be obtained in the first inspection of gas pipeline with positive M$_{res}$.

• 한국지구물리탐사학회:학술대회논문집
• /
• 2006.06a
• /
• pp.175-180
• /
• 2006

The horizontal to vertical ratio technique in spectral domain is a common useful technique to estimate empircal site transfer function. The technique, originally proposed by Nakamura, is porposed to analyse the surface waves in the micortremor records. The purpose of this paper is to estimate spectral ratio using observed data at the seismic stations in Korean Peninsula from the Fukuoka earthquake including many aftershocks. The results show that most of the stations have fairly good amplification factors. However, some of the stations show that very high amplification factors at narrow high frequency band. Those stations which have very high amplification factors seem to do some kind of mechanical consideration for quality observation.

• Journal of Magnetics
• /
• v.20 no.1
• /
• pp.86-90
• /
• 2015

Electromagnetic actuators are versatile and able to meet demanding requirements, such as operation in very low or very high temperatures. When the actuator is used in a high-temperature environment up to $500^{\circ}C$, we need to know how the force-producing capability of the actuator is affected by the operating temperature. Specifically, it is necessary to know the temperature-dependence of magnetic properties that determine the mechanical forces. In this paper, we measure the changes in magnetic properties of SUS410 material in high-temperature environment. We also devise a novel signal processing technique to remove the integration drift. At the field strength of 18,000 A/m, we found that the flux density at $500^{\circ}C$ is decreased by 26%, compared to the result at room temperature. Therefore, the actuator must be sized appropriately, if it is to operate in high-temperature settings.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.5
• /
• pp.454-460
• /
• 2020

Artificial intelligence assistants that provide speech recognition operate through cloud-based voice recognition with high accuracy. In cloud-based speech recognition, Wake-Up-Word (WUW) detection plays an important role in activating devices on standby. In this paper, we compare the performance of Convolutional Neural Network (CNN)-based WUW detection models for mobile devices by using Google's speech commands dataset, using the spectrogram and mel-frequency cepstral coefficient features as inputs. The CNN models used in this paper are multi-layer perceptron, general convolutional neural network, VGG16, VGG19, ResNet50, ResNet101, ResNet152, MobileNet. We also propose network that reduces the model size to 1/25 while maintaining the performance of MobileNet is also proposed.

• Korean Journal of Breeding Science
• /
• v.40 no.4
• /
• pp.435-438
• /
• 2008

A new spray rose cultivar 'Morning Shine' was bred from the cross between light pink spray cultivar 'Charming' and white spray cultivar 'Princess' at the National Horticulture Research Institute. The cross was made in 2000 and 'Morning Shine' was finally selected in 2006 after investigating characteristics for three years from 2003 to 2006. 'Morning Shine', a light yellow spray cultivar grows vigorously and has high productivity. The major characteristics of this cultivar are $159.7 stems/m^2/year$ in yield, 74.3 cm in length of cut flower, 7.2 cm in flower diameter, 35.6 in petal number, and 13.0 days in vase life. This cultivar can be propagated by both cutting and grafting. The consumer's preference of this cultivar is relatively higher than that of control cultivar, 'Sunny Lady'.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.209-209
• /
• 2022

Reservoirs play a key role in the carbon cycle between terrestrial and marine systems and are pathways that release greenhouse gases(GHGs), CO₂, CH₄, and N₂O, into the atmosphere by decomposing organic matters. Developed countries have been actively conducting research on carbon emission assessment of dam reservoirs for over 10 years under the leadership of UNESCO/IHA, but associated research is very rare in Korea. In particular, the GHGs footprint evaluation, which calculates the change in net carbon emission considering the watershed environment between pre- and post- impoundment, is very important in evaluating the carbon emission of hydroelectric dams. The objective of this study was to estimate the GHG emissions and footprints in Daecheong Reservoir using the G-res Tool, an online platform developed by UNESCO/IHA. The G-res Tool estimates CO₂ and CH₄ emissions in consideration of diverse pathway fluxes of GHGs from the reservoir and characterizes changes in GHG fluxes over 100 years based on the expected lifetime of the dam. The input required to use the G-res Tool include data related to watersheds, reservoirs, and dams, and most were collected through the government's public portal. As a result of the study, the GHG footprint of Daecheong Reservoir was estimated to be 93 gCO₂eq/m²/yr, which is similar to that of other reservoirs around the world in the same climate zone. After impoundment, the CH₄ diffusion emission from the reservoir was 73 gCO₂eq/m²/yr, also similar to those of the overseas reservoirs, but the CH₄ bubbling emission, degassing emission, and CO₂ diffusion emissions were 44, 34, 252 gCO₂eq/m²/yr, respectively, showing a rather high tendency. Since the dam reservoir carbon footprint evaluation is essential for the Clean Development Mechanism evaluation of hydroelectric power generation, continuous research is needed in the future. In particular, experimental studies that can replace the emission factors obtained from the overseas dam reservoirs currently used in the G-res Tool should be promoted.

• Korean Journal of Breeding Science
• /
• v.41 no.4
• /
• pp.574-578
• /
• 2009

A new red leaf lettuce (Lactuca sativa L.) cultivar, "Gopungjeokchukmyeon" which has late bolting and good red expression leaf color was developed from a cross between "Andongjeokchukmyeon" (red leaf color and late bolting) and "Dukseomjeokchukmyeon" (high yield). The cross and selection for advanced lines had been done by the pedigree method during 2000-2006. The advanced lines were evaluated for yield and adaptability at several locations in Korea (Gangwon, Kyunggi, Chungbuk, Kyungnam, Jeonnam, Jeonbuk, and Jeju) from 2007 to 2008. The "Gopungjeokchukmyeon" has gray seed color and traverse elliptic leaves. Compared to "Dukseomjeokchukmyeon", marketable yield of "Gopungjeokchukmyeon" was higher by 17% (at 330 g per plant) and Gopungjeokchukmyeon has particularly improved expression of red leaf color in high temperature cultivation in the field. The shelf-life of "Gopungjeokchukmyeon" was three weeks longer than "Dukseomjeokchukmyeon" at $4^{\circ}C$. The anthocyanin and sugar contents of "Gopungjeokchukmyeon" were higher than those of "Dukseomjeokchukmyeon" with 29.4 mg/100 g and 4.8 brix, respectively. The content anthocynin of "Gopungjeokchukmyeon" is about 10 times higher than those of "seomjeokchukmyeon". Furthermore, its taste is better, more crispy, and sweeter than those of "Dukseomjeokchukmyeon", This new cultivar "Gopungjeokchukmyeon" can be harvested all year around even summer cultivation.

• Proceedings of the Korean Powder Metallurgy Institute Conference
• /
• 2006.09b
• /
• pp.1119-1120
• /
• 2006

High Energy Milling (HEM) is applied for the grinding of cement and this can lead to substantial refinement $(<2{\mu}m)$ and mechanically activation of the powder particles. The present paper reviews the preliminary studies, explains the novel technique and suggests the route into commercial application. Particular attention is paid to wear results with an applied $Si_3N_4-grinding$ unit where no substantial wear was found after 4000 h of operation.