이장원 교수

Real-time object detection is crucial for many applications of Unmanned Aerial Vehicles (UAVs) such as reconnaissance and surveillance, search-and-rescue, and infrastructure inspection. In the last few years, Convolutional Neural Networks (CNNs) have emerged as a powerful class of models for recognizing image content, and are widely considered in the computer vision community to be the de facto standard approach for most problems. However, object detection based on CNNs is extremely computationa

Type 5 phosphodiesterase terminates the action of nitric oxide (NO) induced 3',5'-cyclic monophosphate (cGMP). Sildenafil inhibits this phosphodiesterase, increases cellular cGMP concentrations and enhances NO-induced smooth muscle relaxation. We investigated the effect of sildenafil on the oesophageal motor function of healthy subjects and patients with nutcracker oesophagus. Eight healthy volunteers and nine patients with nutcracker oesophagus participated in this study. The participants under

We design a new approach that allows robot learning of new activities from unlabeled human example videos. Given videos of humans executing an activity from their own viewpoint (i.e., first-person videos), our objective is to make the robot learn the temporal structure of the activity as its future regression network, and learn to transfer such model for its own motor execution. We present a new fully convolutional neural network architecture to regress the intermediate scene representation corr

Robot learning from demonstration (LfD) is a research paradigm that can play an important role in addressing the issue of scaling up robot learning. Since this type of approach enables non-robotics experts can teach robots new knowledge without any professional background of mechanical engineering or computer programming skills, robots can appear in the real world even if it does not have any prior knowledge for any tasks like a new born baby. There is a growing body of literature that employ Lf

We present a first step towards developing an interactive piano tutoring system that can observe a student playing the piano and give feedback about hand movements and musical accuracy. In particular, we have two primary aims: 1) to determine which notes on a piano are being played at any moment in time, 2) to identify which finger is pressing each note. We introduce a novel two-stream convolutional neural network that takes video and audio inputs together for detecting pressed notes and finger

Rotating machines, such as pumps and compressors, are critical components in refinery and chemical plants used to transport fluids between processing units. Bearings are often the critical parts of rotating machinery, and their failure could result in economic loss and/or safety issues. Therefore, estimation of the remaining useful life (RUL) of a bearing plays an important role in reducing production losses and avoiding machine damage. Because bearing failure mechanisms tend to be complex and s

Computer vision techniques that can anticipate people»s actions ahead of time could create more responsive and natural human-robot interaction systems. In this paper, we present a new human gesture forecasting framework for human-drone interaction. Our primary motivation is that despite growing interest in early recognition, little work has tried to understand how people experience these early recognition-based systems, and our human-drone forecasting framework will serve as a basis for conducti

The vertically aligned carbon nanotube (CNT)-bundle is utilized as a mechanical buffer for inertial micro-switch to extend the contact time and therefore to obtain reliable and stable output signals. The CNT bundles are directly synthesized on two facing surfaces of both movable- and stationary electrodes by patterning the catalyst on released microstructures and the following thermal chemical vapor deposition process. When the movable electrode collides with the stationary electrode, CNT bundle

In the last few decades, spectroscopic techniques such as near-infrared (NIR) spectroscopy have gained wide applications in several industries, such as the pharmaceutical, agricultural, oil, and gas industries. As a result, various soft sensors have been developed to predict sample properties from spectroscopic readings. Because the spectroscopic readings at different wavelengths, especially at the adjacent wavelengths, are highly correlated, it has been shown that variable selection could signi

This paper presents multi-domain topology optimization using a current-carrying coil and a ferromagnetic material in order to focus the pulsed magnetic field (PMF) onto a target area. The multi-domain design sensitivities in the harmonic magnetic field are calculated using the adjoint variable method (AVM). In numerical examples, the multi-domain topology optimization technique is applied to the designs of a C-core actuator in the magnetostatic field and a PMF generator in the time-harmonic magn

We design a new approach that allows robot learning of new activities from unlabeled human example videos. Given videos of humans executing the same activity from a human's viewpoint (i.e., first-person videos), our objective is to make the robot learn the temporal structure of the activity as its future regression network, and learn to transfer such model for its own motor execution. We present a new deep learning model: We extend the state-of-the-art convolutional object detection network for

본 연구는 압축공기포 소화설비의 소화성능을 평가하기 위해 포헤드 설비를 이용하여 실험을 진행 하였다. 압축공기포 소화설비는 포수용액에 압축공기를 혼입하여 포를 발생시키는 방식으로 해외에서는 원거리 방수가 가능하고 물 사용량을 줄여 수손피해를 최소화할 수 있는 압축공기포 소화설비(CAFS: Compressed Air Foam System)가 많이 활용되고 있다. 본 연구에서는 UL162 기준으로 수성막포 3 % 포 소화약제를 적용하여 기존의 공기 혼입 방식에 의한 포 소화설비와 압축공기포 소화설비간 비교 실험을 통하여 소화 성능 효과를 비교 분석하였다. 압축공기포 소화설비의 공기 혼입 비율은 포 수용액과 1 : 1의 부피 비율로 하였으며 발포유량은 각각 140 L/min, 160 L/min, 180 L/min, 200 L/min으로 변화를 주면서 소화효과를 검증하였다. 그 결과 소화 성능면에서는 압축공기포 소화설비가 공기 혼입 방식보다 모든 유량 조건에서 소화시간이 빠르게 나타났다.

본 연구는 압축공기를 포수용액(수성막포 3%)에 주입하여 발포하는 압축공기포 소화설비에서 가압 공기 혼입 비율에 따른 소화성능을 평가하고자 하였다. 실험장치는 캐나다국립연구소 및 UL162 기준을 준용하여 제작한 압축공기포전용의 포소화설비 실험장치를 활용하였으며, 소화모형은 소화약제의 형식승인 및 제품검사의 기술기준에 의한 유류화재(B급) 20단위 모형을 적용하였다. 압축공기는 공기혼합기를 통하여 주입하였으며, 공기포비를 1 : 4, 1 : 7, 1 : 10으로 증가시키면서 경향성을 연구하였다. 또한 공기포비 1 : 4에서 합성계면활성제포와 수성막포의 비교 실험도 함께 진행하였다. 소화 성능실험 결과 동일한 방출유량 조건에서 수성막포는 공기포비 1 : 7에서 소화효과가 가장 빠르게 나타났으며 공기포비 1 : 10에서 가장 소화시간이 길게 나타났다. 또한 수성막포와 합성계면활성제포간 비교 실험에서는 수성막포의 소화효과가 더 빠르게 나타났다. This research is to eval