• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.28 no.3
• /
• pp.117-123
• /
• 2016

One of the major characteristics of the swell-like high waves, which occur in the East Sea mostly in winter with large height and long period, is its suddenness associated with the rapid development of high waves from a calm state of sea. To represent such suddenness, in this study, the term sudden high waves is introduced. To propose the criterion of sudden high waves, comparisons were made between the wave measurement data at Gangneung and Wangdolcho for eight years from 2005 and the record of marine accidents and property damage on the coast of Gangwon-do Province and Gyeongsangbuk-do Province during the same period. It was found that most of the accidents occurred when ${\Delta}(H^2L)/{\Delta}t$ was approximately greater than the top 20% or $88.6m^3/hr$, which is therefore proposed as the criterion of sudden high waves. The used variable represents the rate of increase of the wave energy in one wavelength, including not only height and period but also suddenness of high waves.

• Journal of the Korean GEO-environmental Society
• /
• v.17 no.5
• /
• pp.5-16
• /
• 2016

A rainfall induced slope failure is a common natural hazard in mountainous areas worldwide. Sudden and rapid failures which have a high possibility of occurrence in a steep slope are always the most dangerous due to their suddenness and high velocities. Based on a series of experiments this study aimed to determine a critical angle which could be considered as an approximate threshold for a sudden failure. The experiments were performed using 0.42 mm mean grain size sand in a 200 cm long, 60 cm wide and 50 cm deep rectangular flume. A numerical model was created by integrating a 2D seepage flow model and a 2D slope stability analysis model to predict the failure surface and the time of occurrence. The results showed that, the failure mode for the entire material will be sudden for slopes greater than $67^{\circ}$; in contrast the failure mode becomes retrogressive. There is no clear link between the degree of saturation and the mode of failure. The simulation results in considering matric suction showed good matching with the results obtained from experiment. A subsequent discarding of the matric suction effect in calculating safety factors will result in a deeper predicted failure surface and an incorrect predicted time of occurrence.

• Journal of Multimedia Information System
• /
• v.8 no.3
• /
• pp.159-166
• /
• 2021

COVID-19 has dramatically changed people's daily life. Wearing masks is considered as a simple but effective way to defend the spread of the epidemic. Hence, a real-time and accurate mask wearing detection system is important. In this paper, a deep learning-based mask wearing detection system is developed to help people defend against the terrible epidemic. The system consists of three important functions, which are image detection, video detection and real-time detection. To keep a high detection rate, a deep learning-based method is adopted to detect masks. Unfortunately, according to the suddenness of the epidemic, the mask wearing dataset is scarce, so a mask wearing dataset is collected in this paper. Besides, to reduce the computational cost and runtime, a simple online and real-time tracking method is adopted to achieve video detection and monitoring. Furthermore, a function is implemented to call the camera to real-time achieve mask wearing detection. The sufficient results have shown that the developed system can perform well in the mask wearing detection task. The precision, recall, mAP and F1 can achieve 86.6%, 96.7%, 96.2% and 91.4%, respectively.