• Tunnel and Underground Space
• /
• v.29 no.6
• /
• pp.508-522
• /
• 2019

Because of the slope failure, not only property damage but also human damage can occur, slope stability analysis should be conducted to predict and reinforce of the slope. This paper, defines the ground areas that can be characterized in terms of slope failure such as Rockmass jointset, Rockmass fault, Soil, Leakage water and Crush zone in sloped images. As a result, it was shown that the deep learning instance segmentation network can be used to recognize and automatically segment the precise shape of the ground region with different characteristics shown in the image. It showed the possibility of supporting the slope mapping work and automatically calculating the ground characteristics information of slopes necessary for decision making such as slope reinforcement.

• Journal of the Korean Geotechnical Society
• /
• v.38 no.7
• /
• pp.39-47
• /
• 2022

This study aims at evaluating the stability of disaster risks, such as schools, apartments, and geotechnical structures located around slopes in urban areas. The research conducted an aerial photography analysis on where the slope of the retaining wall behind 𐩒𐩒 High School in Gwangju collapsed in August 2018 due to heavy rain. In general, the overflow of rainwater has been managed through drainage channels around slopes during the rainy season, and the surface flow of rainfall was limited due to the presence of dense forests in the area. However, when the slope collapsed, a lot of water flowed out of the ground, and the saturated surface layer ground was destroyed. To analyze the cause, the changed terrain of the upper slope area, which could not be directly identified, was photographed using unmanned aerial vehicles. Digital Elevation Model by unmanned aerial vehicle shooting was performed by analyzing the slope map, calculating the direction of rainfall and the length and width of water-logged areas. The change in the instability of the slope over time due to a 10-day rainfall was also analyzed through numerical analysis.

• The Journal of Engineering Geology
• /
• v.28 no.3
• /
• pp.489-499
• /
• 2018

A flume experiments was used to study the characteristics of the surface displacements and volumetric water contents (VWC) during torrential rain. The surface displacement and VWC of the granite weathered soil were measured for rainfall intensity (100, 200 mm/hr) and initial ground condition (VWC 7, 14, 26%). The test processes were also recorded by video cameras. According to the test results, The shallow failure is classified into three types: retrogressive failure, progressive failure and defined failure. In the case of retrogressive failure and progressive failure, relatively large damage could occur due to the feature that soil is deposited to the bottom of the slope. the shallow failure occurred when the VWC reached a certain value regardless of the initial soil condition. It was found that the shallow failure can be predicted through the increase patton of the VWC under the condition of the ground dry condition (VWC 7%) and the natural condition (VWC 14%). For high rainfall intensity, progressive failure predominated, and rainfall intensity above a certain level did not affect wetting front transition.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.156-157
• /
• 2018

본 논문에서 제안하는 기술은 무인기 획득 영상에 dense optical flow 기술을 적용하여, 이미지 내에서 급격히 움직이는 사물을 추출하는 기술이다. 제안 기술을 활용하여 사람, 사물 장소에 해당하는 시간에 따른 데이터인 다중로그 데이터로 융합 분석하여 낙석, 산사태, 비탈면 붕괴등을 감지 할 수 있게 되어 보행자의 안전을 보장 하고자 한다. 본 논문에서는 해당 기술을 구현하기 위한 무인기 및 탑재 장비와 데이터 처리를 위한 서버들간의 인터페이스 및 분석 알고리즘을 소개한다.