• 한국멀티미디어학회논문지
• /
• 제22권11호
• /
• pp.1269-1279
• /
• 2019

In medical image analysis, segmentation is considered as a vital process since it partitions an image into coherent parts and extracts interesting objects from the image. In this paper, we consider automatic segmentations of OCT retinal images to find six layer boundaries using convolutional neural networks. Segmenting retinal images by layer boundaries is very important in diagnosing and predicting progress of eye diseases including diabetic retinopathy, glaucoma, and AMD (age-related macular degeneration). We applied well-known CNN architecture for general image segmentation, called Segnet, U-net, and CNN-S into this problem. We also proposed a shortest path-based algorithm for finding the layer boundaries from the outputs of Segnet and U-net. We analysed their performance on public OCT image data set. The experimental results show that the Segnet combined with the proposed shortest path-based boundary finding algorithm outperforms other two networks.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제30권7_8호
• /
• pp.719-726
• /
• 2003

본 논문에서는 Visible Human 컬러 단면 영상에서 인접한 지방 영역과 색상 특성이 유사하여 구별이 매우 힘든 뼈 영역을 분할하기 위해 다중 모달 정합 방법을 제안한다. 뼈와 그 인접영역의 구별이 뚜렷한 CT 영상에서 뼈를 분할하고 두 영상의 정합을 이용하여 컬러 영상에서 최종 뼈 분할을 수행한다. CT 영상에서 뼈의 분할 방법은 임계값 기반 방법을 사용하였고, 정합은 두 영상에서 신체 부위를 임계값 기반의 방법을 사용하여 분할된 객체들의 경계를 상호 상관관계(cross-correlation)방법을 사용하여 수행하였다. 제안된 방법은 Visible Human 컬러 단면 영상 중에 뼈와 인접 지방이 유사하여 그 분할이 어려운 머리부위와 다리부위에 적용하여 고무적인 결과론 얻었다.

• 한국주조공학회지
• /
• 제29권1호
• /
• pp.33-37
• /
• 2009

Cartesian grid system has mainly been used in the casting simulation even though it does not nicely represent sloped and curved surfaces. These distorted boundaries cause several problems. A special treatment is necessary to clear these problems. In this paper, we propose a new method that can consider the cutting cells which are cut by casting and mold based on the partial cell treatment (PCT). This method provides a better representation of geometry surface and will be used in the computation of velocities that are defined on the cell boundaries in the Cartesian grid system. Various test examples for several casting process were computed and validated. The analysis results of more accurate fluid flow pattern and less momentum loss owing to the stepped boundaries in the Cartesian grid system were confirmed. By using the cut cell method, performance of computation gets better because of reducing the whole number of meshes.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제37권11호
• /
• pp.866-870
• /
• 2010

본 논문에서는 근접 에지라는 새로운 에너지 항을 추가한 개선된 스네이크 알고리즘을 제안한다. 제안된 알고리즘은 스네이크 셀 주위에 근접하는 에지가 있을 경우 이 에지와 스네이크 셀 간의 거리를 에너지로 나타내고, 이 에너지를 전체 에너지 함수에 포함시킴으로써 물체의 윤곽선 탐색을 보다 효과적으로 수행한다. 근접 에지 기반의 스네이크 알고리즘은 셀이 물체의 오목한 경계 부분으로 탐색하는 것을 가능하게 하며, 에너지 항 사이에 실험적인 가중치 조정을 거치지 않고도 복잡한 물체의 윤곽선을 강인하게 검출할 수 있다. 제안된 방법의 성능 평가를 위한 실험에서는 개선된 스네이크 알고리즘이 속도를 크게 저하시키지 않으면서 윤곽선 추출의 정확도를 보다 개선하였음을 확인할 수 있었다.

• 한국재료학회:학술대회논문집
• /
• 한국재료학회 2011년도 춘계학술발표대회
• /
• pp.239.1-239.1
• /
• 2011

Dye sensitized solar cells (DSC) are promising devices for inexpensive, nontoxic, transparent, and large-scale solar energy conversion. Generally thick $TiO_2$ nanoporous films act as efficient photoanodes with their large surface area for absorbing light. However, electron transport through nanoparticle networks causes the slowdown and the loss of electron transport because of a number of interparticle boundaries inside the conduction path. We have studied DSCs with precisely dimension-controlled $TiO_2$ nanotubes array as photoanode. $TiO_2$ nanotubes array is prepared by template-directed fabrication method with atomic layer deposition. Well-ordered nanotubes array provides not only large surface area for light absorbing but also direct pathway for electrons with minimalized grain boundaries. Large enlongated anatase grains in the nanotubes could enhance the conductivity of electrons, but also suppress the recombination with holes through defect sites during diffusion into the electrode. To study the effect of grain boundaries, we fabricated two kinds of nanotubes which have different grain sizes by controlling deposition conditions. And we studied electron conduction through two kinds of nanotubes with different grain structures. The solar cell performance was studied as a function of thickness and grain structures. And overall solar-to-electric energy conversion efficiencies of up to 7% were obtained.

• 한국CDE학회논문집
• /
• 제4권3호
• /
• pp.173-179
• /
• 1999

In reverse engineering, when a shape containing multi-patched surfaces is digitized, the boundaries of these surfaces should be detected. The objective of this paper is to introduce a computationally efficient segmentation technique for extracting edges, ad partitioning the 3D measuring point data based on the location of the boundaries. The procedure begins with the identification of the edge points. An automatic edge-based approach is developed on the basis of local geometry. A parametric quadric surface approximation method is used to estimate the local surface curvature properties. the least-square approximation scheme minimizes the sum of the squares of the actual euclidean distance between the neighborhood data points and the parametric quadric surface. The surface curvatures and the principal directions are computed from the locally approximated surfaces. Edge points are identified as the curvature extremes, and zero-crossing, which are found from the estimated surface curvatures. After edge points are identified, edge-neighborhood chain-coding algorithm is used for forming boundary curves. The original point set is then broke down into subsets, which meet along the boundaries, by scan line algorithm. All point data are applied to each boundary loops to partition the points to different regions. Experimental results are presented to verify the developed method.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 하계학술대회 논문집 D
• /
• pp.1361-1363
• /
• 1998

Poly-Si grain boundaries act as potential barriers as well as recombination centers for the photo-generated carriers in solar cells. Thereby, grain boundaries of poly-Si are considered as a major source of the poly-Si cell efficiency was reduced This paper investigated grain boundary effect of poly-Si wafer prior to the solar cell fabrication. By comparing I-V characteristics inner grain, on and across the grain boundary, we were able to detect grain potentials. To reduce grain boundary effect we carried out pretreatment, $POCl_3$ gettering, and examined carrier lifetime. This paper focuses on resistivity variation effect due to grain boundary of poly-Si. The resistivity of the inner grain was $2.2{\Omega}-cm$, on the grain boundary$2.3{\Omega}-cm$, across the grain boundary $2.6{\Omega}-cm$. A measured resistivity varied depending on how many grains were included inside the four point probes. The resistivity increased as the number of grain boundaries increased. Our result can contribute to achieve high conversion efficiency of poly-Si solar cell by overcoming the grain boundary influence.

• 한국레이저가공학회지
• /
• 제7권3호
• /
• pp.25-36
• /
• 2004

This study was performed to clarification of the formation of weld discontinuities in the dissimilar laser fusion zone. Welding parameters were beam power of 1300, 1430, 1560, and 1700 W and travel speed of 1, 1.3, and 1.7 m/min. Most cavities in the fusion zone were observed near the tip. Cavities in the fusion zone observed to be formed and grown from pores in the tip. More cavities were formed as the beam position moves to the tip side. Small cavities were decreased but large cavities were increased when the energy input increased. W content in the fusion zone was increased with heat input and as the beam position close to the tip. In the fusion zone, W content in the dendrite boundary was increased with heat input. Considering the propagation path and fracture morphology, cracks were solidification cracking, and were initiated and propagated along the dendrite boundaries. The formation of cracks might be related with the W rich ${\mu}$ phase which was formed in the grain boundaries and dendrite boundaries.

• 한국음향학회지
• /
• 제18권8호
• /
• pp.9-16
• /
• 1999

일반적으로 사람은 말을 할 때 어절들은 몇몇의 구로 그룹핑하여 발음함으로써 발화한다. 이것은 듣는 사람으로 하여금 발화의 의미와 의도를 잘 파악하도록 도와준다. 특히, 이러한 목적으로 발화자는 무의식적으로 운율정보(억양, 장단, 리듬 등)를 적절히 사용하게 된다. 본 논문에서는 발화된 문장에서 운율경계를 인식의 단위로 하는 음성인식방법에 대하여 제안한다. 즉, 발화된 문장을 운율구단위로 나누는 방법을 제안하고 나누어진 단위에 따라 연속음 인식실험을 수행하였다. 인식실험결과 연속음인식 시간의 감소를 관찰할 수 있었으며, 물론 음성인식률도 20-10%정도 증가하였다.

• 말소리와 음성과학
• /
• 제9권4호
• /
• pp.9-17
• /
• 2017

The aim of this study is to examine whether Chinese Korean learners (CKL) can correctly produce the aspiration in 'a lenis obstruents /k/, /t/, /p/, /ʧ/+/h/ sound' sequence at the lexical and post-lexical level. For this purpose 4 Korean native speakers (KNS), 10 advanced and 10 intermediate CKL participated in a production test. The material analyzed consisted of 10 Korean sentences in which aspiration can be applied at different prosodic boundaries (syllable, word, accentual phrase). The results showed that for KNS and CKL, the rate of application of aspiration was different according to prosodic boundaries. Aspiration was more frequently applied at the lexical level than at the post-lexical level and it was more frequent at the word boundary than at the accentual phrase boundary. For CKL, pronunciation errors were either non-application of aspiration or coda obstruent omission. In the case of non-application of aspiration, CKL produced the target syllable as an underling form and they did not transform it as a surface form. In the case of coda obstruent ommision, most of the errors were caused by the inherent complexity of phonological process.