• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.9
• /
• pp.2199-2205
• /
• 2013

In modern society the image processing has been applied to various digital devices such as smartphone, digital camera, and digital TV. In the field of image processing the edge detection is one of the important parts in the image processing procedure. The image edge means point that the pixel value is changed between background and object rapidly, and includes the important information such as magnitude, location, and orientation. The performance of the existing edge detection method is insufficient for the image degraded by AWGN(additive white Gaussian noise) because it detects edges by using small weighted masks. Therefore, in this paper, to detect edge in AWGN environment effectively, we proposed an algorithm that detects edge as calculated gradient of sorting vector which is transformed by estimated mask from new pixel according to each region.

• Korean Chemical Engineering Research
• /
• v.46 no.3
• /
• pp.465-472
• /
• 2008

Fabrication of nanostructured materials and synthesis of nanomaterials have intensively studied to realize electronic devices for nanotechnology. By using nanoporous alumina mask, nanostructured material can be fabricated in the form of uniform array. The size and the density of the nanostructured materials can be controllable by changing the pore diameter and the density of the alumina mask. This method is possible low cost and on large scale process, and feasible to contribute the fusion technology consisting of information technology, nanotechnology, and biotechnology. Therefore, these techniques provide alternative approaches for development of new electronic applications. In this paper, the fabrication technique and its applications of nanoporous alumina mask are described and nanostructured materials such as quantum dots, nanoholes, and nanorods are introduced.

• Proceedings of the Korean Vacuum Society Conference
• /
• 1999.07a
• /
• pp.57-57
• /
• 1999

공기 플라즈마로 다이아몬드 박막을 식각하여 한 방향으로 정렬된 고밀도의 다이아몬드 바늘을 만들었다. 다이아몬드 기판에 적당한 양의 몰리브데늄 박막을 입힌 후 수백 볼트의 음 전압을 걸고 RF 또는 직류 플라즈마로 식각하였다. 바늘의 모양과 밀도는 기판 온도, 압력, 바이어스 전압, 전력, 몰리브데늄 양 등의 식각조건에 따라 결정되었다. 기판 온도가 높으면 바늘이 굵어졌다. 몰리브데늄 박막은 식각마스크로 작용하였는데 때때로 몰리브데늄으로 만들어진 기판 장착대가 스퍼트링되어 저절로 몰리브데늄 박막이 입혀져서 바늘이 만들어지기도 하여 바늘 밀도, 형상을 정밀하게 조절하기 위해서는 적당한 양의 몰리브데뉴 박막을 미리 입히는 것이 좋다. 조건이 잘 맞으면 굵기가 0.1$\mu\textrm{m}$, 바늘 사이 간격이 0.1$\mu\textrm{m}$, 높이는 3$\mu\textrm{m}$ 이상인 다이아몬드 바늘 격자가 만들 수 있었다. 이러한 바늘은 다결정 다이아몬드 박막 뿐만 아니라 고온고압 다이아몬드, 자연 다이아몬드 등으로부터도 만들 수 있었다. 다이아몬드 바늘은 전계 전자 방출 소자, 복합 재료를 위한 다이아몬드 섬유, 방열판, 다공질 다이아몬드 등으로 사용할 수 있다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2004.05b
• /
• pp.627-629
• /
• 2004

본 연구는 자기정렬 방법을 기존의 방식과 다르게 적용하여 수소화 된 비정질 실리콘 박막 트랜지스터의 제조공정을 단순화하고, 박막 트랜지스터의 게이트와 소오스-드레인간의 기생용량을 줄인다. 본 연구의 수소화 된 비정질 실리콘 박막 트랜지스터는 Inverted Staggered 형태로 게이트 전극이 하부에 있다 실험 방법은 게이트전극, 절연층, 전도층, 에치스토퍼 및 포토레지스터층을 연속 증착한다. 스토퍼층을 게이트 전극의 패턴으로 남기고, 그 위에 n+a-Si:H 층 및 NPR(Negative Photo Resister)을 형성시킨다. 상부 게이트 전극과 반대의 패턴으로 NPR층을 패터닝하여 그것을 마스크로 상부 n+a-Si:H 층을 식각하고, 남아있는 NPR층을 제거한다. 그 위에 Cr층을 증착한 후 패터닝 하여 소오스-드레인 전극을 위한 Cr층을 형성시켜 박막 트랜지스터를 제조한다. 이렇게 제조하면 기존의 박막 트랜지스터에 비하여 특성은 같고, 제조공정은 줄어들며, 또한 게이트와 소오스-드레인간의 기생용량이 줄어들어 동작속도를 개선시킬 수 있다.

• Journal of the Korean Institute of Telematics and Electronics D
• /
• v.36D no.5
• /
• pp.21-28
• /
• 1999

A new method of making high speed self-aligned ESD (Elevated Source/Drain) MOSFET is proposed. Different from the conventional LDD (Lightly-Doped Drain) structure, the proposed ESD structure needs only one ion implantation step for the source/drain junctions, and makes it possible to modify the depth of the recessed channel by use of dry etching process. This structure alleviates hot-carrier stress by use of removable nitride sidewall spacers. Furthermore, the inverted sidewall spacers are used as a self-aligning mask to solve the self-align problem. Simulation results show that the impact ionization rate ($I_{SUB}/I_{D}$) is reduced and DIBL (Drain Induced Barrier Lowering) characteristics are improved by proper design of the structure parameters such as channel depth and sidewall spacer width. In addition, the use of removable nitride sidewall spacers also enhances hot-carrier characteristics by reducing the peak lateral electric field in the channel.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2002.10a
• /
• pp.141-144
• /
• 2002

Computer code was developed to design waveguide gratings based on coupled-mode equations and the transfer matrix formalism. The experimental set-up has been constructed for inscribing Bragg gratings in planar waveguides with a phase mask and uv laser beam, which enables alignment and packaging of grating devices as well as in-situ performance measurements. Bragg grating has been fabricated on silica planar waveguides with 0.75% Germanium concentration and its transmittance spectrum was measured to have 95% reflectance at the peak wavelength. Optical losses as the function of the misalignment were measured and their usage is described.

• Proceedings of the KAIS Fall Conference
• /
• 2001.05a
• /
• pp.269-272
• /
• 2001

실리콘 마이크로머시닝 기술과 바이폴라 공정으로 집적화된 압력센서를 제작하고 동작특성 평가를 수행하였다. 센서부 보상파라미터를 추출하였고 트리밍 공정을 통하여 출력전압의 보상을 수행하였다. 센서 특성은 압저항 위치, 마스크 정렬 오차, 다이어프램 정밀두께제어 정도, 보호막의 과도식각 정도 등에 의하여 민감하게 좌우됨을 알 수 있었다. 웨이퍼별 샘플추출을 통하여 센서부 감도는 평균 0.653mV/kPa, 감도의 온도계수는 -2078.8ppm/℃, 옵셋 전압은 30.78mV, 옵셋전압의 온도계수는 32.11㎶/℃로 측정되었다. 추출된 샘플의 다이어프램 두께오차는 27±2.5㎛였다. 센서부 특성평가 결과를 통하여 신호처리회로의 옵셋 및 스팬보상, 온도보상을 위한 트리밍 공정을 수행한 결과 개발사양을 만족하는 결과를 얻을 수 있었다.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.7
• /
• pp.335-347
• /
• 2024

Open-vocabulary 3D point cloud instance segmentation (OV-3DIS) is a challenging visual task to segment a 3D scene point cloud into object instances of both base and novel classes. In this paper, we propose a novel model Open3DME for OV-3DIS to address important design issues and overcome limitations of the existing approaches. First, in order to improve the quality of class-agnostic 3D masks, our model makes use of T3DIS, an advanced Transformer-based 3D point cloud instance segmentation model, as mask proposal module. Second, in order to obtain semantically text-aligned visual features of each point cloud segment, our model extracts both 2D and 3D features from the point cloud and the corresponding multi-view RGB images by using pretrained CLIP and OpenSeg encoders respectively. Last, to effectively make use of both 2D and 3D visual features of each point cloud segment during label assignment, our model adopts a unique feature ensemble method. To validate our model, we conducted both quantitative and qualitative experiments on ScanNet-V2 benchmark dataset, demonstrating significant performance gains.

• Journal of KIISE
• /
• v.42 no.10
• /
• pp.1222-1230
• /
• 2015

Segmentation of the anterior cruciate ligament (ACL) in knee MRI remains a challenging task due to its inhomogeneous signal intensity and low contrast with surrounding soft tissues. In this paper, we propose a multi-atlas-based segmentation of the ACL in knee MRI with locally-aligned probabilistic atlas (PA) in an iterative graph cuts framework. First, a novel PA generation method is proposed with global and local multi-atlas alignment by means of rigid registration. Second, with the generated PA, segmentation of the ACL is performed by maximum-aposteriori (MAP) estimation and then by graph cuts. Third, refinement of ACL segmentation is performed by improving shape prior through mask-based PA generation and iterative graph cuts. Experiments were performed with a Dice similarity coefficients of 75.0%, an average surface distance of 1.7 pixels, and a root mean squared distance of 2.7 pixels, which increased accuracy by 12.8%, 22.7%, and 22.9%, respectively, from the graph cuts with patient-specific shape constraints.

• Convergence Security Journal
• /
• v.14 no.5
• /
• pp.17-24
• /
• 2014

This study proposed LDP (Local Directional Pattern) as a new local micro pattern for facial expression recognition to solve noise sensitive problem of LBP (Local Binary Pattern). The proposed method extracts 8-directional components using $m{\times}m$ mask to solve LBP's problem and choose biggest k components, each chosen component marked with 1 as a bit, otherwise 0. Finally, generates a pattern code with bit sequence as 8-directional components. The result shows better performance of rotation and noise adaptation. Also, a new local facial feature can be developed to present both PFF (permanent Facial Feature) and TFF (Transient Facial Feature) based on the proposed method.