• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2006년도 춘계학술대회 논문집
• /
• pp.177-178
• /
• 2006

Many authors issued the feature-preserving averaging technique according to positioning and scaling process using landmarks, which represent the geometric characteristics of three dimensional surface models. Such a technique should be done by manual procedure, choosing and marking the landmarks on each bone surface before averaging process. In this study, we produced another averaging technique without having to use such manual procedure, and made averaging models from three dimensional surface data that were reconstructed from computerized tomography images of Digital Korean Project. The bone models were subjected to orthogonal coordinator system. These models were transformed to coincide mass center and to align principal axis. Then, bone models were scaled according to average length data of sample bone models on all axis(x, y, z). After establishing voxellar hexahedron space which contain all sample bone models, we counted the number of overlapping for each voxel. We generated the three dimensional average surface by displaying the yokels that have more overlapping number than boundary number. The boundary number was decided when the average volume of each bone equal to the volume of bone that would be averaged. Using this technique, we can make a feature-preserving averaging volume of bones.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1134-1139
• /
• 2003

Tracking is a fundamental technique which is able to be applied to gesture recognition, visual surveillance, tangible agent and so forth. Especially, multiple object tracking has been extensively studied in recent years in order to perform many and more complicated tasks. In this paper, we propose a new approach of multiple object tracking which is based on discrete event. We call this system the DEMOT (Discrete Event based Multiple Object Tracking). This approach is based on the fact that a multiple object tracking can have just four situations - initiation, continuation, termination, and overlapping. Here, initiation, continuation, termination, and overlapping constitute a primary event set and this is based on the change of the number of extracted objects between a previous frame and a current frame. This system reduces computational costs and holds down the identity of all targets. We make experiments for this system with respect to the number of targets, each event, and processing period. We describe experimental results that show the successful multiple object tracking by using our approach.

• 한국지능시스템학회논문지
• /
• 제22권1호
• /
• pp.94-99
• /
• 2012

본 논문은 대규모 비선형 시스템의 분산 샘플치 제어 문제를 논한다. 대상 시스템은 타카기-수게노 퍼지모델로 표현된다. 중복분해 기법을 이용하여 아날로그 분산 제어기를 설계하고 이것은 지능형 디지털 재설계 기법에 의하여 샘플치 제어기로 변환된다. 설계조건은 선형행렬부등식으로 표현된다. 모의실험 결과에 의하여 제안된 기법의 효용성을 입증한다.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2006년도 춘계학술대회논문집
• /
• pp.850-853
• /
• 2006

This paper presents a primary study for a new method of enhancing measurement using impedance analysis method, which is widely used for structure health monitoring. Many of skyscrapers, bridges, tunnels, which have been built up several years, and nearly reached their operating life due to aging crack, material deterioration, and so on. therefore, the health monitoring technique has obtained considerable attention. The usage of the PZT sensor is proposed, and the coupling torque of bolt-connection overlapping beam is assumed. By using the torque wrench, it is easy to apply torque of different values to each bolt. The coupling torque is adjusted gradually from $3N{\cdot}m\;to\;0N{\cdot}m$ with the decrement of $1N{\cdot}m$. Simultaneously, the impedance peak variation is measured. The peak frequency variation can be determined through the change of bolt coupling torque.

• Journal of Minimally Invasive Surgery
• /
• 제21권4호
• /
• pp.160-167
• /
• 2018

Purpose: The aim of our study was to present an abdominal wall closure technique using barbed suture $V-Loc^{TM}$ 90 after single incision laparoscopic appendectomy (SILA) and to compare perioperative outcomes with conventional layer by layer abdominal wall closure after SILA. Methods: From March 2014 to July 2016, a retrospective case-control study was conducted for a total of 269 consecutive patients who underwent SILA. According to abdominal wall closure methods, 129 patients were classified into the V-Loc closure group and 140 patients were assigned into the conventional layer by layer closure group. In the V-Loc group, abdominal wall closure was performed from the fascia to the skin with a single thread of unidirectional absorbable barbed suture $V-Loc^{TM}$ 90 2-0 using continuous running suture and reverse overlapping reinforced running technique. Subcutaneous closure and subcuticular suture were performed with the remaining portion of V-Loc. Results: The V-Loc closure group showed shorter total operation time ($40.0{\pm}15.4min$ vs. $44.9{\pm}16.3min$, p=0.013) and abdominal wall cusing continuous running suture and reverse overlapping reinforced running technique. Subcutaneous closure and subcuticular suture were performed with the remaining portion of V-Loc. Results: The V-Loc closure group showed shorter total operation time losure time ($5.5{\pm}0.9min$ vs. $6.5{\pm}0.8min$, p<0.001). Postoperative incision length was significantly shorter in the V-Loc closure group ($1.1{\pm}0.3cm$ vs. $1.8{\pm}0.4cm$, p<0.001). Postoperative wound pain, time to resume diet, postoperative hospital stay, complications including surgical site infection, or mean patient satisfaction score at one month after hospital discharge was not significantly different between the two groups. Conclusion: In conclusion, unidirectional knotless barbed suture is a safe alternative method for abdominal wall closure after SILA. It can save time while providing comparable cosmesis.

• 복식
• /
• 제58권9호
• /
• pp.1-17
• /
• 2008

One of features of fashion in Joseon period is culture of overlapping of several clothes. This culture appeared with Hapbong(合縫:sewing of several clothes as one) at late Joseon period and developed into simpler and more practical fashion culture. Hapbong is mainly identified at men's official uniform such as Jaebok(祭服), Jobok(朝服), Kwanbok(官福), Gugunbok(具軍服), and women's formal dress such as Wonsam(圓衫) and Dangeui(唐衣) which formed several clothes get to be a set. Such fashion was designed to maintain dignity and power while simple to wear. And It has the trace of overlapping visually on the collar, sleeve, breast-tie and others. Hapbong is attributable to thoughts of practical science, post-toadyism, modernization, renovated the system of clothing, and advanced sewing technique.

• 국제학술발표논문집
• /
• The 4th International Conference on Construction Engineering and Project Management Organized by the University of New South Wales
• /
• pp.445-452
• /
• 2011

The schedule hierarchy in construction project is generally composed of three levels. The highest level is a milestone schedule and represented by Bar Chart format. The middle level is an integrated project schedule (IPS) and represented by CPM (Critical Path Method) format. The lowest level is a detail working schedule and usually represented by Bar Chart. The traditional scheduling techniques such as ADM (Arrow Diagramming Method) or PDM (Precedence Diagramming Method) cannot represent all kinds of schedule within schedule hierarchy as identical schedule format. However, the BDM (Beeline Diagramming Method) technique can represent all kinds of schedule within schedule hierarchy as identical CPM format. This paper describes the basic concept, principle, interpretation methods, and schedule computation methods of the BDM as a new networking technique that can represent all kinds of overlapping relationships between activities, and then presents an example of representing three level's schedules within schedule hierarchy by the BDM technique.

• 한국건설관리학회논문집
• /
• 제17권3호
• /
• pp.71-80
• /
• 2016

대규모 복합 프로젝트(complex project) 또는 메가 프로젝트(mega project)들에서 확률적으로 공정계획을 수립하는 대표적인 방법이 PERT (Project Evaluation and Review Technique)기법이다. PERT기법은 작업공기의 불확실성에만 중점을 두고 작업간 연결관계는 Finish-to-Start(FS)관계로 제한되기 때문에 작업간 중복관계가 존재할 경우 작업간 연결관계의 불확실성에 대한 확률적인 추정이 불가능하다. PERT기법의 이러한 한계를 극복하기 위해, 작업간 다양한 중복관계들의 불확실성을 고려하여 사업전체 공기를 추정하는 방법으로 PLET (Probabilistic Linkage Evaluation Technique)기법을 제안되었다. 그러나 PLET 기법에 의한 사업공기 추정은 작업간 연결관계의 불확실성만을 고려할 뿐, PERT기법에서의 작업공기에 대한 불확실성은 고려하지 않으므로, 완전한 사업공기 추정방법이라 할 수 없다. 본 논문은 작업공기의 불확실성을 고려하는 PERT기법과 작업간 연결관계의 불확실성을 고려하는 PLET기법을 결합시켜 사업공기를 추정하는 통합 모델을 제안하고 확률적 타당성을 검증한다. 그리고 통합 모델의 일정계산방법을 제시하고 이를 실제사례에 적용하여 검증함으로써, 고위험 건설사업의 공정계획을 수립할 때 보다 폭넓고 다양한 위험요인들을 통합적으로 고려하는 확률적 사업공기 추정 방법을 제안하는 것이다.

• 대한의용생체공학회:학술대회논문집
• /
• 대한의용생체공학회 1997년도 춘계학술대회
• /
• pp.72-74
• /
• 1997

We have developed a 3-D image processing and display technique that include image resampling, modification of MIP, and fusion of MIP image and volumetric rendered image. This technique facilitates the visualization of the three-dimensional spatial relationship between vasculature and surrounding organs by overlapping the MIP image on the volumetric rendered image of the organ. We applied this technique to a MR brain image data to produce an MRI angiogram that is overlapped with 3-D volume rendered image of brain. MIP technique was used to visualize the vasculature of brain, and volume rendering was used to visualize the other structures of brain. The two images are fused after adjustment of contrast and brightness levels of each image in such a way that both the vasculature and brain structure are well visualized either by selecting the maximum value of each image or by assigning different color table to each image. The resultant image with this technique visualizes both the brain structure and vasculature simultaneously, allowing the physicians to inspect their relationship more easily. The presented technique will be useful for surgical planning for neurosurgery.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제10권7호
• /
• pp.191-198
• /
• 2021

분산 딥러닝은 각 노드에서 지역적으로 업데이트한 지역 파라미터를 동기화는 과정이 요구된다. 본 연구에서는 분산 딥러닝의 효과적인 파라미터 동기화 과정을 위해, 레이어 별 특성을 고려한 allreduce 통신과 연산 오버래핑(overlapping) 기법을 제안한다. 상위 레이어의 파라미터 동기화는 하위 레이어의 다음 전파과정 이전까지 통신/계산(학습) 시간을 오버랩하여 진행할 수 있다. 또한 이미지 분류를 위한 일반적인 딥러닝 모델의 상위 레이어는 convolution 레이어, 하위 레이어는 fully-connected 레이어로 구성되어 있다. Convolution 레이어는 fully-connected 레이어 대비적은 수의 파라미터를 가지고 있고 상위에 레이어가 위치하므로 네트워크 오버랩 허용시간이 짧고, 이를 고려하여 네트워크 지연시간을 단축할 수 있는 butterfly all-reduce를 사용하는 것이 효과적이다. 반면 오버랩 허용시간이 보다 긴 경우, 네트워크 대역폭을 고려한 ring all-reduce를 사용한다. 본 논문의 제안 방법의 효과를 검증하기 위해 제안 방법을 PyTorch 플랫폼에 적용하여 이를 기반으로 실험 환경을 구성하여 배치크기에 대한 성능 평가를 진행하였다. 실험을 통해 제안 기법의 학습시간은 기존 PyTorch 방식 대비 최고 33% 단축된 모습을 확인하였다.