• Journal of the Korean Society of Propulsion Engineers
• /
• v.5 no.1
• /
• pp.49-59
• /
• 2001

Functionally gradient materials(FGMs) generally consist of different material components, such as ceramics and metals. Continuous changes in their composition, microstructure, porositys, and so on result in gradients in the properties of FGMs. In this study AlN/Al-$Al_3$Ti FGM cubes were fabricated by the powder metallurgical process, and the characteristics of the FGMs were investigated. Powders of Mg and Ti mixed with Al in different ratios or the stack of the mixed powders having a stepwisely controlled compositional gradient from one plane to another were formed in a steel mold. The more the composition of Mg and sintering temperature are increased, the more the amount of AlN is formed. AlN and $Al_3$Ti distributed with continuously gradient composition were detected by XRD analysis and also revealed in optical microstructures, and microhardness tests.

• Investigative Magnetic Resonance Imaging
• /
• v.18 no.2
• /
• pp.87-106
• /
• 2014

Purpose : To describe how a robust implementation of a radial 3D gradient-echo sequence with stack-of-stars sampling can be achieved, to review the imaging properties of radial acquisitions, and to share the experience from more than 5000 clinical patient scans. Materials and Methods: A radial stack-of-stars sequence was implemented and installed on 9 clinical MR systems operating at 1.5 and 3 Tesla. Protocols were designed for various applications in which motion artifacts frequently pose a problem with conventional Cartesian techniques. Radial scans were added to routine examinations without selection of specific patient cohorts. Results: Radial acquisitions show significantly lower sensitivity to motion and allow examinations during free breathing. Elimination of breath-holding reduces failure rates for non-compliant patients and enables imaging at higher resolution. Residual artifacts appear as streaks, which are easy to identify and rarely obscure diagnostic information. The improved robustness comes at the expense of longer scan durations, the requirement for fat suppression, and the nonexistence of a time-to-center value. Care needs to be taken during the configuration of receive coils. Conclusion: Routine clinical use of radial stack-of-stars sequences is feasible with current MR systems and may serve as substitute for conventional fat-suppressed T1-weighted protocols in applications where motion is likely to degrade the image quality.

• Geophysics and Geophysical Exploration
• /
• v.2 no.1
• /
• pp.33-42
• /
• 1999

AVO and complex trace analyses mainly used to characterize natural gas reservoir were tested in this paper for a possible application to detection of major geological discontinuities such as fracture zones. The test data used in this study were calculated by utilizing a viscoelastic numerical program which was based on the generalized Maxwell body for a horizontal fracture model. In AVO analysis of a horizontal fracture zone, p-wave reflection appears to be variant depending upon the acoustic-impedence contrast and the offset distance. The fracture zone is also effectively clarified both in gradient stack and range-limited stack in which fracture zone reflection is attenuated with the increasing offset distance. In complex attribute plots (instantaneous amplitude, frequency, and phase), the top and bottom of the fracture Tone are characterized by a zone of strong amplitudes and an event of the same phase. Low frequency characteristics appear at the fracture zone and the underneath. Amplitude attenuation and waveform dispersion are dependent on Q-contrast between the fracture zone and the surrounding media. They were properly compensated by optimum inverse Q-filtering.

• Transactions on Electrical and Electronic Materials
• /
• v.18 no.4
• /
• pp.181-184
• /
• 2017

Multilayer ceramics in which piezoelectric layers of $0.90Pb(Zr_{0.48}Ti_{0.52})O_3-0.05Pb(Mn_{1/3}Sb_{2/3})O_3-0.05Pb(Zn_{1/3}Nb_{2/3})O_3$ (0.90PZT-0.05PMS-0.05PZN) stack alternately with silver electrode layers were prepared by an advanced low-temperature co-fired ceramic (LTCC) method. The electrical properties and bonding strength of the multilayers were associated with the interface morphologies between the piezoelectric and silver-electrode layers. Usually, the inner silver electrodes are fabricated by sintering silver paste in multi-layer stacks. To improve the interface bonding strength, piezoelectric powders of 0.90PZT-0.05PMS-0.05PZN with an average particle size of $23{\mu}m$ were added to silver paste to form a gradient interface. SEM observation indicated clear interfaces in multilayer ceramics without powder addition. With the increase of piezoelectric powder addition in the silver paste, gradient interfaces were successfully obtained. The multilayer ceramics with gradient interfaces present greater bonding strength as well as excellent piezoelectric properties for 30~40 wt% of added powder. On the other hand, over addition greatly increased the resistance of the inner silver electrodes, leading to a piezoelectric behavior like that of bulk ceramics in multilayers.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.543-545
• /
• 2019

본 논문은 최신 한국어 의존 구문 분석 모델(Korean dependency parsing model)들과 다양한 앙상블 모델(ensemble model)들을 결합하여 그 성능을 분석한다. 단어 표현은 미리 학습된 워드 임베딩 모델(word embedding model)과 ELMo(Embedding from Language Model), Bert(Bidirectional Encoder Representations from Transformer) 그리고 다양한 추가 자질들을 사용한다. 또한 사용된 의존 구문 분석 모델로는 Stack Pointer Network Model, Deep Biaffine Attention Parser와 Left to Right Pointer Parser를 이용한다. 최종적으로 각 모델의 분석 결과를 앙상블 모델인 Bagging 기법과 XGBoost(Extreme Gradient Boosting) 이용하여 최적의 모델을 제안한다.

• Knowledge Management Research
• /
• v.23 no.3
• /
• pp.1-22
• /
• 2022

This study aims to investigate the gamification effect of the badge awards, the most popular gamification process, on users participation behavior. This study also attempts to investigate the effect of tailored gamification, which designs the system of gamification differently based on users' characteristics, focusing on the level of online user information disclosure. For this, we collect and analyze data on 557 users and 1,048,020 answers from StackOverflow, an online Q&A community for developers. The results show that providing a badge is effective for increasing the amount of user participation, whereas providing a goal through the badge is partially effective for increasing the quality of participation. However, the moderating effect of whether users disclose their SNS information on the relationship between badge gaining and participation decrease is not statistically significant. For platform operators, our findings emphasize the importance of gamification design to enhance user engagement effectively.

• Korean Journal of Remote Sensing
• /
• v.33 no.4
• /
• pp.391-400
• /
• 2017

Geo-based application services linked with the Open Geospatial Consortium (OGC) Web Processing Service (WPS) protocol have been regarded as an important standardized framework for of digital earth building in the web environments. The WPS protocol provides interface standards for analysis functionalities within geo-spatial processing in web-based service systems. Despite its significance, there is few performance tests of WPS applications. The main motivation in this study is to perform the comparative performance test on WPS standards. Test system, which was composed of WPS servers, WPS framework, data management module, geo-based data processing module and client-sided system, was implemented by fully open source stack. In this system, two kinds of geo-based image processing functions such as cloud detection and gradient magnitude computation were applied. The performance test of different server environments of non-WPS, synchronous WPS 1.0 and asynchronous WPS 2.0 was carried out using 100 threads and 400 threads corresponds client users on a web-based application service. As the result, at 100 threads, performance of three environments was within an adjacent range in the average response time to complete the processing of each thread. At 400 threads, the application case of WPS 2.0 showed the distinguished characteristics for higher performance in the response time than the small threads cases. It is thought that WPS 2.0 contributes to settlement of without performance problems such as time delay or thread accumulation.

• Journal of the Optical Society of Korea
• /
• v.14 no.4
• /
• pp.431-437
• /
• 2010

A stack of gradient-index (GRIN) rod lenses cannot be used for coherent anti-Stokes Raman scattering (CARS) microendoscopy for insertion to internal organs through a surgical keyhole with minimal invasiveness. That's because GRIN lens has large amount of inherent chromatic aberrations in spite of absolutely requiring a common focus for pump and Stokes beam with each frequency of ${\omega}_p$ and ${\omega}_S$. For this endoscopic purpose, we need to develop a long slender probe-type objective, namely probe-type microscope objective (PMO). In this paper, we introduce the structure, the working principle, and the design techniques of PMO which is composed of a probe-type lens module (PLM) and an adaptor lens module (ALM). PLM is first designed for a long slender type and ALM is successively designed by using several design parameters from PLM for eliminating optical discords between scanning unit and PLM. A combined module is optimized again to eliminate some coupling disparities between PLM and ALM for the best PMO. As a result, we can obtain a long slender PMO with perfectly diffraction-limited performance for pump beam of 817 nm and Stokes beam of 1064 nm.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.142-146
• /
• 2023

We present a method for generating 3D structures and rendering objects by combining VAE (Variational Autoencoder) and GAN (Generative Adversarial Network). This approach focuses on generating and rendering 3D models with improved quality using residual learning as the learning method for the encoder. We deep stack the encoder layers to accurately reflect the features of the image and apply residual blocks to solve the problems of deep layers to improve the encoder performance. This solves the problems of gradient vanishing and exploding, which are problems when constructing a deep neural network, and creates a 3D model of improved quality. To accurately extract image features, we construct deep layers of the encoder model and apply the residual function to learning to model with more detailed information. The generated model has more detailed voxels for more accurate representation, is rendered by adding materials and lighting, and is finally converted into a mesh model. 3D models have excellent visual quality and accuracy, making them useful in various fields such as virtual reality, game development, and metaverse.

• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.8
• /
• pp.307-324
• /
• 2018

Recently, recurrent neural networks have been attracting attention in solving prediction problem of sequential data through structure considering time dependency. However, as the time step of sequential data increases, the problem of the gradient vanishing is occurred. Long short-term memory models have been proposed to solve this problem, but there is a limit to storing a lot of data and preserving it for a long time. Therefore, research on memory-augmented neural network (MANN), which is a learning model using recurrent neural networks and memory elements, has been actively conducted. In this paper, we describe the structure and characteristics of MANN models that emerged as a hot topic in deep learning field and present the latest techniques and future research that utilize MANN.