• 한국방사선학회논문지
• /
• 제11권7호
• /
• pp.655-661
• /
• 2017

방사선 치료 전 환자 위치 확인을 위해 수행하는 콘빔 CT 촬영에서 환자 선량 감소를 위해 Sparse view CT가 사용되고 있다. 본 연구는 시뮬레이션과 실험을 통해 선형보간법과 inpainting 방법을 이용하여 사이노그램의 sparse 데이터 복원하고 평가하는 것이다. 사이노그램 복원은 여러 간격의 각도로 획득된 영상에 적용되었다. 복원된 사이노그램은 역투영재구성법으로 재구성되었고, 그 결과를 평균제곱근오차와 영상의 프로파일로 나타내었다. 결과에 따르면, 평균제곱근오차와 영상 프로파일은 투영 각도와 복원법에 의존하였다. 시뮬레이션과 실험 결과에서 inpainting 복원법은 선형보간법에 비해 사이노그램의 복원 측면에서 개선된 결과를 보여주었다. 따라서, inpainting 방법은 환자 선량을 감소시키면서 영상화질을 유지시키는데 기여할 수 있을 것이다.

• 정보보호학회논문지
• /
• 제28권1호
• /
• pp.85-93
• /
• 2018

생체 정보를 이용한 사용자 인증은 차세대 인증 방법으로서 기존의 인증 시스템에서 급진적으로 사용되고 있는 인증 방법이다. 대부분의 생체 인증 시스템은 수집된 생체 정보가 가지는 노이즈로 인한 문제, 데이터의 품질에 대한 문제, 인식률의 한계 등 많은 문제점들을 가지고 있다. 이를 해결하기 위한 방법으로 본 논문에서는 비선형적인 실제 데이터를 정확하게 처리하기 위해 비선형기법인 Dual QML을 사용하고, 또한 정확한 영역을 추출하여 인증의 정확도를 증가시키는 전처리 과정을 추가로 제안하여 정확도 증가뿐만 아니라 성능을 향상시키는 방법을 제안하고자 한다. 앞서 발표된 Dual QML은 생체 정보로 얼굴, 장문, 귀를 사용하였다. 본 논문은 앞선 Dual QML 실험에 사용하지 않은 홍채를 생체 정보로 사용하여 홍채 인식을 위한 방법으로도 Dual QML이 우수하다는 것을 보이고자 한다. 마지막으로 실험을 통해 이에 대한 실증을 보이고자 한다.

• 한국통신학회논문지
• /
• 제29권3C호
• /
• pp.419-425
• /
• 2004

본 논문은 카메라로 찍은 투사 사영(Perspective Projection)의 한 영상으로부터 물체의 크기와 위치를 계산하기 위해서 3 개의 소실 점만을 사용해서 계산하는 새로운 방법을 제안한다. 우리의 접근 방법은 투사 사영의 영상으로부터 초점 거리(Focal Length), 회전 행렬(Rotation Matrix) 등의 정보들 없이 3개의 소실 점만을 가지고 계산하는 방법이다. 물체는 꼭지점(vertices)의 좌표가 크기 벡터 v 의 선형 함수로서 표현할 수 있는 다각형으로써 모델이 된다는 것을 가정한다. 이 재구성의 입력은 영상에서 특징 점과 모델에서 특징 점 사이 대응점의 집합으로 표현할 수 있다. 이 매개 변수 모델의 각각 크기를 최적화하기 위해서, 재구성의 최적화는 동일하게 매개 변수 공간을 샘플링에 의한 최적화기에 대하여 다중 시작점(multiple starting points)을 발생하는 다중 시작(multi-start) 방법을 가지는 표준 비선형 최적화 기법을 효과적으로 해결할 수가 있다.

• Archives of Plastic Surgery
• /
• 제34권4호
• /
• pp.490-492
• /
• 2007

Purpose: Nipple reconstruction is an important step in breast reconstruction after mastectomy. There are considerable number of reconstructive methods developed over the past years. Each of these has not only its own special advantages, but also limitations. Therefore, no single method has become the overwhelming favorite. Sometimes it seems to be compromised when the nipple must be located directly over a linear scar. Methods: A 48-year-old female patient received a central lumpectomy with circumareolar resection of the nipple areolar complex 4 months ago. The newly designed nipple must be positioned directly astride a scar. We drew two equal-sized rectangular flaps sharing a common limb on a transverse scar and the result was two opposing plow form. Each flap size was about 1.3 cm wide and 2.5 cm long. First we elevated the flap from the distal part at a deep dermal plane, then deepened the level of dissection to raise the dermal-fat flaps. The donor site could be closed directly without any dog-ear deformity. Then we folded down the elevated flaps and loosely sutured skin with nonabsorbable materials. Each flap inner side was approximated side by side. Finally we made new natural nipple with 6 mm projection. We applied tattooing in the areola area with micropigmentation device after three months. Results: After ten months of follow-up periods, the nipple projection was stable and symmetric. The nipple projection was 3.1 mm, compared with 2.8 mm for the opposite nipple. Conclusion: Our experiences shows that this double opposing plow flap is a particularly useful and simple technique when there is a traverse scar crossing the center of the proposed nipple area.

• 대한전자공학회논문지SP
• /
• 제40권3호
• /
• pp.203-216
• /
• 2003

In this paper a new feature extraction and target classification method is proposed for the recognition part of FLIR(Forwar Looking Infrared)-image-based ATR system. Proposed feature extraction method is "cluster(=set of classes)-based"version of previous fisherfaces method that is known by its robustness to illumination changes in face recognition. Expecially introduced class clustering and cluster-based projection method maximizes the performance of fisherfaces method. Proposed target image classification method is based on the mixture of experts model which consists of RBF-type experts and MLP-type gating networks. Mixture of experts model is well-suited with ATR system because it should recognizee various targets in complexed feature space by variously mixed conditions. In proposed classification method, one expert takes charge of one cluster and the separated structure with experts reduces the complexity of feature space and achieves more accurate local discrimination between classes. Proposed feature extraction and classification method showed distinguished performances in recognition test with customized. FLIR-vehicle-image database. Expecially robustness to pixelwise sensor noise and un-wanted intensity variations was verified by simulation.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• 제1권1호
• /
• pp.74-82
• /
• 2020

Predictions of suitable habitat areas can provide important information pertaining to the risk assessment and management of alien plants at early stage of their establishment. Here, we predict the invasion potential of Muhlenbergia capillaris (pink muhly) in South Korea using five bioclimatic variables. We adopt four models (generalized linear model, generalized additive model, random forest (RF), and artificial neural network) for projection based on 630 presence and 600 pseudo-absence data points. The RF model yielded the highest performance. The presence probability of M. capillaris was highest within an annual temperature range of 12 to 24℃ and with precipitation from 800 to 1,300 mm. The occurrence of M. capillaris was positively associated with the precipitation of the driest quarter. The projection map showed that suitable areas for M. capillaris are mainly concentrated in the southern coastal regions of South Korea, where temperatures and precipitation are higher than in other regions, especially in the winter season. We can conclude that M. capillaris is not considered to be invasive based on a habitat suitability map. However, there is a possibility that rising temperatures and increasing precipitation levels in winter can accelerate the expansion of this plant on the Korean Peninsula.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4502-4518
• /
• 2015

Fisher linear discriminant analysis (LDA) is one of the most popular projection techniques for feature extraction and has been widely applied in face recognition. However, it cannot be used when encountering the single sample per person problem (SSPP) because the intra-class variations cannot be evaluated. In this paper, we propose a novel method called local similarity based linear discriminant analysis (LS_LDA) to solve this problem. Motivated by the "divide-conquer" strategy, we first divide the face into local blocks, and classify each local block, and then integrate all the classification results to make final decision. To make LDA feasible for SSPP problem, we further divide each block into overlapped patches and assume that these patches are from the same class. To improve the robustness of LS_LDA to outliers, we further propose local similarity based median discriminant analysis (LS_MDA), which uses class median vector to estimate the class population mean in LDA modeling. Experimental results on three popular databases show that our methods not only generalize well SSPP problem but also have strong robustness to expression, illumination, occlusion and time variation.

• 대한전자공학회논문지SP
• /
• 제45권6호
• /
• pp.64-69
• /
• 2008

얼굴영상의 인식 기술은 보안과 감시를 비롯하여 머신 인터페이스와 콘텐츠 검색 등에서 활용이 광범위 하다. 그러나 주로 고해상도 영상이 연구의 대상이었고 원거리에서 획득된 저해상도 표적에 대하여 상대적으로 드물게 연구가 이루어졌다. 본 논문에서는 포톤 카운팅(Photon-counting) 선형판별법을 이용하여 저해상도 환경에서 얼굴영상의 인식을 수행한다. 포톤 카운팅 선형판별법은 Fisher 선형 판별법에서 발생하는 특이행렬 문제없이 Fisher의 최적화 기준을 실현한다. 즉, 차원의 축소나 특징 추출 과정 없이 고차원 공간에서 최적화된 투영을 위한 선형판별함수를 구성하고 이를 이용하여 판정하므로 저해상도 환경을 비롯한 얼굴영상의 왜곡의 극복에 효과적이다. 실험 결과는 제안한 방법이 주성분 분석을 활용하는 Eigen face 또는 주성분 분석과 Fisher 선형판별법이 결합된 Fisher face보다 우수하다는 것을 보여준다.

• Nuclear Engineering and Technology
• /
• 제45권7호
• /
• pp.895-910
• /
• 2013

The reflood test data from the rod bundle heat transfer (RBHT) test facility showed that the grids in the upper portion of the rod bundle could become wet well before the arrival of the quench front and that the sizes of liquid droplets downstream of a wet grid could not be predicted by the droplet breakup models for a dry grid. To investigate the water droplet generation from a wet grid spacer, a viscous linear temporal instability model of the water sheet issuing from the trailing edge of the grid with the surrounding steam up-flow is developed in this study. The Orr-Sommerfeld equations along with appropriate boundary conditions for the flow are solved using Chebyshev series expansions and the Tau-Galerkin projection method. The effects of several physical parameters on the water sheet oscillation are studied by determining the variation of the temporal growth rate with the wavenumber. It is found that a larger relative steam velocity to water velocity has a tendency to destabilize the water sheet with increased dynamic pressure. On the other hand, a larger ratio of steam boundary layer to the half water sheet thickness has a stabilizing effect on the water sheet oscillation. Droplet diameters downstream of the spacer grid predicted by the present model are found to compare reasonably well with the data obtained at the RBHT test facility as well as with other data recently reported in the literature.

• 한국농공학회논문집
• /
• 제54권5호
• /
• pp.141-153
• /
• 2012

In this study, we estimated probable precipitation amounts at the target year (2020, 2030, 2040) of 55 weather stations in Korea using the 24 hour annual maximum precipitation data from 1973 through 2009 which should be useful for management of agricultural reservoirs. Not only trend tests but also non-stationary tests were performed and non-stationary frequency analysis were conducted to all of 55 sites. Gumbel distribution was chosen and probability weighted moment method was used to estimate model parameters. The behavior of the mean of extreme precipitation data, scale parameter, and location parameter were analyzed. The probable precipitation amount at the target year was estimated by a non-stationary frequency analysis using the linear regression analysis for the mean of extreme precipitation data, scale parameter, and location parameter. Overall results demonstrated that the probable precipitation amounts using the non-stationary frequency analysis were overestimated. There were large increase of the probable precipitation amounts of middle part of Korea and decrease at several sites in Southern part. The non-stationary frequency analysis using a linear model should be applicable to relatively short projection periods.