• 대한임베디드공학회논문지
• /
• 제16권3호
• /
• pp.107-111
• /
• 2021

A 3D point cloud map is an essential elements in various fields, including precise autonomous navigation system. However, generating a 3D point cloud map using a single sensor has limitations due to the price of expensive sensor. In order to solve this problem, we propose a precise 3D mapping system using low-cost sensor fusion. Generating a point cloud map requires the process of estimating the current position and attitude, and describing the surrounding environment. In this paper, we utilized a commercial visual-inertial odometry sensor to estimate the current position and attitude states. Based on the state value, the 2D LiDAR measurement values describe the surrounding environment to create a point cloud map. To analyze the performance of the proposed algorithm, we compared the performance of the proposed algorithm and the 3D LiDAR-based SLAM (simultaneous localization and mapping) algorithm. As a result, it was confirmed that a precise 3D point cloud map can be generated with the low-cost sensor fusion system proposed in this paper.

• Smart Structures and Systems
• /
• 제24권6호
• /
• pp.759-768
• /
• 2019

In nuclear fusion reactors, the key structural component (i.e., the plasma-facing component) undergoes high heat flux cyclic loading. To ensure the safety of fusion reactors, an experimental study on the temperature-induced creep of stainless steel under heat flux cyclic loading was performed in the present work. The strains were measured using a stereo digital image correlation technique (3D-DIC). The influence of the heat haze was eliminated, owing to the use of a vacuum environment. The specimen underwent heat flux cycles ($500^{\circ}C-1000^{\circ}C$) with different mechanical preloads (0 kN, 10 kN, 30 kN, and 50 kN). The results revealed that, for a relatively large preload (for example, 50 kN), a single temperature cycle can induce a residual strain of up to $15000{\mu}{\varepsilon}$.

• 정보보호학회논문지
• /
• 제16권3호
• /
• pp.115-127
• /
• 2006

본 논문의 목적은 다중 생체 인식을 위하여 사용되는 다양한 정규화함수와 결합 및 패턴 분류 알고리즘들의 성능을 비교 평가하는 것이다. 이를 위하여 NIST에서 제공하는 유사도 집합인 BSSR(Biometric from Set-Releasel) 데이터베이스와 다양한 정규화함수, 결합 및 패턴 분류 알고리즘을 이용하여 실험을 수행하였으며, HTER(Half Total Error Rate)을 이용한 평가 결과를 제시하고 있다. 본 연구는 단일 데이터베이스와 평가 항목을 이용한 평가 결과를 제시함으로써 다중 생체 인식시스템의 성능 개선 연구를 위한 토대가 될 수 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권10호
• /
• pp.3390-3405
• /
• 2022

Long Short-Term Memory (LSTM) combined with attention mechanism is extensively used to generate semantic sentences of images in image captioning models. However, features of salient regions and spatial information are not utilized sufficiently in most related works. Meanwhile, the LSTM also suffers from the problem of underutilized information in a single time step. In the paper, two innovative approaches are proposed to solve these problems. First, the Synergy-Gated Attention (SGA) method is proposed, which can process the spatial features and the salient region features of given images simultaneously. SGA establishes a gated mechanism through the global features to guide the interaction of information between these two features. Then, the Recurrent Fusion LSTM (RF-LSTM) mechanism is proposed, which can predict the next hidden vectors in one time step and improve linguistic coherence by fusing future information. Experimental results on the benchmark dataset of MSCOCO show that compared with the state-of-the-art methods, the proposed method can improve the performance of image captioning model, and achieve competitive performance on multiple evaluation indicators.

• 한국수소및신에너지학회논문집
• /
• 제32권4호
• /
• pp.219-227
• /
• 2021

Hydrogen isotopes, which are used as raw materials in fusion reaction, participate in the reaction only in small amount, and most of them are released together with impurities. In order to recover and reuse only hydrogen isotopes from this exhaust gas, a recovery process is required, and most of the hydrogen isotopes can be recovered using a Pd Membrane. In this study, the recovery rate of hydrogen isotopes was measured through the first and second stage Pd membrane experiments. In the case of the experiment using a single stage Pd membrane, about 99.2%, and in the case of the first stage and second stage Pd membrane connection experiments, a recovery rate of 99.9% or more was obtained. Therefore, the recovery rate of Pd membrane process applied to hydrogen can be applied to hydrogen isotopes. In addition, the simulation model was established using aspen custom modeler, a commercial software, and the validity of the simulation was checked by applying the references and experimental data. The simulation results based on the experimental data showed a difference of 2% or less.

• Journal of Korean Neurosurgical Society
• /
• 제60권6호
• /
• pp.691-700
• /
• 2017

Objective : The authors prospectively analyzed the effect of one-level or two-level anterior cervical discectomy and fusion (ACDF), comparing stand-alone cages and cage-with-plate fixation constructs with respect to clinical outcomes and radiologic changes. Methods : A total of 84 patients who underwent one-level (n=52) or two-level ACDF (n=32) for cervical disc disease and who completed 2 years of follow-up were included in this study. The patients were divided by cervical level and grouped into ACDF-Cage-only and ACDF-Cage-with-plate groups. The following parameters were assessed using radiographs : subsidence, C2-C7 lordosis angle, fusion segment angle, adjacent disc space narrowing, and fusion status. Clinical outcomes were assessed using the neck disability index (NDI) and visual analog scale scores for arm pain. Results : In the comparison of one-level ACDF-cage-only and ACDF-cage-with-plate groups, the NDI score was better in the cage-only group at the 3-, 12-, and 24-month follow-ups : however, no significant difference in clinical outcomes was observed. In the comparison of two-level ACDF-cage-only and ACDF-cage-with-plate groups, no difference in any clinical outcome was observed between the two groups. At the 24-month follow-up, subsidence was observed in 45.8% of patients in the one-level cage-only group and 32.1% of patients in the one-level cage-with-plate fixation group. There was no statistically significant difference in the incidence rate between the two groups (p=0.312). Subsidence in the two-level cage-only group (66.6%) was significantly more frequent than in the two-level cage-with-plate fixation group (30%; p=0.049). The fusion rate for patients in the one-level cage-only group was not significantly different from that in the one-level cage-with-plate fixation group (cage-only, 87.5%; cage-with-plate fixation, 92.9%; p=0.425) ; fusion rate in the two-level patients were also similar between groups (cage-only, 83.3%; cage-with-plate fixation, 95%; p=0.31). Conclusion : Our clinical results showed that for single-level cases, plate fixation had no additional benefit versus cage-only; for two-level ACDF cases, the fusion rate and clinical outcomes were similar, although the cage-with-plate fixation group had a lower incidence of cage subsidence than did the cage-only group. We conclude that physicians should be aware of this possible disadvantage associated with using cervical plates in one-level ACDF. However, in two-level ACDF, subsidence is more likely to occur without plate fixation, and thus the addition of plate fixation should be considered.

• 대한전자공학회논문지SP
• /
• 제49권4호
• /
• pp.77-84
• /
• 2012

얼굴인식 기술은 지능형 보안, 웹에서 콘텐츠 검색, 지능로봇의 시각부분, 머신인터페이스 등, 활용이 광범위 하다. 그러나 일반적으로 대상자의 표정과 포즈 변화, 주변의 조명 환경과 같은 문제가 있으며 이와 더불어 원거리에서 획득한 영상의 경우 저해상도를 비롯하여 블러와 잡음에 의한 영상의 열화 등의 여러 가지 어려움이 발생한다. 본 논문에서는 포톤 카운팅(Photon-counting) 선형판별법(Linear Discriminant Analysis)을 이용한 다중 분류기(Classifier)에 의한 판정을 융합하여 얼굴 영상 인식을 수행한다. Fisher 선형판별법은 집단 간 분산을 최대로 하고 집단 내 분산을 최소로 하는 공간으로 선형 투영하는 방법으로, 학습영상의 수가 적을 경우 특이행렬 문제가 발생하지만 포톤카운팅 선형 판별법은 이러한 문제가 없으므로 차원축소를 위한 전 처리 과정이 필요 없다. 본 논문의 다중 분류기는 포톤 카운팅 선형판별법의 유클리드 거리(Euclidean Distance) 또는 정규화된 상관(Normalized Correlation)을 적용하는 판정규칙에 따라 구성된다. 다중분류기의 판정의 융합은 각 분류기 cost의 정규화(Normalization), 유효화(Validation), 그리고 융합규칙(Fusion Rule)으로 구성된다. 각 분류기에서 도출된 cost는 같은 범위로 정규화된 후 유효화 과정에서 선별되고 Minimum, 또는 Average, 또는 Majority-voting의 융합규칙에 의하여 융합된다. 실험에서는 원거리에서 획득한 효과를 구현하기 위하여 고해상도 데이터베이스 영상을 인위적으로 Unfocusing과 Motion 블러를 이용하여 열화하여 테스트하였다. 실험 결과는 다중분류기 융합결과의 인식률은 단일분류기보다 높다는 것을 보여준다.

• Asian Spine Journal
• /
• 제12권6호
• /
• pp.1010-1016
• /
• 2018

Study Design: Retrospective cohort study. Purpose: To report on spinal fusion assessment using computed tomography (CT) after adult spinal deformity (ASD) surgery using ultra-low dose recombinant human bone morphogenetic protein 2 (RhBMP-2). Overview of Literature: The reported dose of RhBMP-2 needed for successful spinal posterolateral fusion in ASD ranges from 10 to 20 mg per spinal level. This study reports the use of ultra-low dose of RhBMP-2 (0.07 mg per facet) to achieve spinal fusion in multilevel ASD surgery. Methods: Consecutive patients who underwent ASD surgery using ultra-low dose RhBMP-2 were recruited. Routine postoperative CT analysis for spinal fusion was performed by two spine surgeons. Inter-observer agreement was calculated for facet fusion (FF) and interbody fusion (IBF) at 6 and 12 months after the procedure. Results: Six consecutive ASD patients with a mean age of 62 years (28-72 years) were examined. Each patient received a total dose of 12 mg with an average dose of $0.69{\pm}0.2mg$ (0.42-1 mg) per single FF and $1.38{\pm}0.44mg$ (0.85-2 mg) for IBF. Total 131 FF and 15 IBF were examined in the study, with 88 FFs and nine IBFs being analyzed specifically at 6 months after the surgery. FF and IBF reported by surgeons A and B at 6 months were 97.7% vs. 91.9% FF, respectively (${\kappa}=0.95$) and 100% vs. 100% IBF, respectively (${\kappa}=1$). Two patients underwent longitudinal follow-up CT at 12 months, and the FF rates reported by surgeons A and B were 100% vs. 95.8%, respectively (${\kappa}=0.96$). Five out of nine facet (56%) non-unions were identified at the cross-links. The remaining four facet pseudarthrosis were noted at 1-2 spinal levels caudal to the cross-links. At the final clinical follow-up, there was no rod breakage, deformity progression, neurological deficit, or symptom recurrence. The Oswestry Disability Index improved by an average of $32.8{\pm}6.3$, while the mental component summary of the 36-item Short-Form Health Survey improved by an average of $4.7{\pm}2.1$, and physical component summary improved by an average of $10.5{\pm}2.1$. Conclusions: To our knowledge, this is the first study to report a CT that defined 92%-98% FF and 100% IBF using the lowest reported dose of RhBMP-2 in multilevel ASD surgery. The use of ultra-low dose RhBMP-2 reduces the RhBMP-2 related complications and healthcare costs.

• Journal of Microbiology and Biotechnology
• /
• 제14권3호
• /
• pp.643-645
• /
• 2004

A $\beta$-xylosidase was purified from the culture of Trichoderma sp. SY. The ten-day-old culture filtrate was concentrated, followed by ion-exchange chromatography and gel filtration chromatography. As a result, $\beta$-xylosidase was finally purified about 53-fold and appeared as a single band by SDS-PAGE. The optimum pH and temperature were 5.0 and $55^{\circ}C$, respectively, and the molecular weight about 80 kDa. The purified $\beta$-xylosidase was found to be inhibited by various metal ions but not inhibited by xylose.

• Journal of information and communication convergence engineering
• /
• 제15권1호
• /
• pp.7-13
• /
• 2017

We consider a distributed estimation where many nodes remotely placed at known locations collect the measurements of the parameter of interest, quantize these measurements, and transmit the quantized data to a fusion node; this fusion node performs the parameter estimation. Noting that quantizers at nodes should operate in a non-regular framework where multiple codewords or quantization partitions can be mapped from a single measurement to improve the system performance, we propose a low-weight estimation algorithm that finds the most feasible combination of codewords. This combination is found by computing the weighted sum of the possible combinations whose weights are obtained by counting their occurrence in a learning process. Otherwise, tremendous complexity will be inevitable due to multiple codewords or partitions interpreted from non-regular quantized data. We conduct extensive experiments to demonstrate that the proposed algorithm provides a statistically significant performance gain with low complexity as compared to typical estimation techniques.