• Journal of Information Processing Systems
• /
• v.11 no.1
• /
• pp.22-38
• /
• 2015

While Modern Standard Arabic is the formal spoken and written language of the Arab world; dialects are the major communication mode for everyday life. Therefore, identifying a speaker's dialect is critical in the Arabic-speaking world for speech processing tasks, such as automatic speech recognition or identification. In this paper, we examine two approaches that reduce the Universal Background Model (UBM) in the automatic dialect identification system across the five following Arabic Maghreb dialects: Moroccan, Tunisian, and 3 dialects of the western (Oranian), central (Algiersian), and eastern (Constantinian) regions of Algeria. We applied our approaches to the Maghreb dialect detection domain that contains a collection of 10-second utterances and we compared the performance precision gained against the dialect samples from a baseline GMM-UBM system and the ones from our own improved GMM-UBM system that uses a Reduced UBM algorithm. Our experiments show that our approaches significantly improve identification performance over purely acoustic features with an identification rate of 80.49%.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.54-54
• /
• 2000

For its synergistic relationship, an integrated SDINS/GPS system has been adopted in many navigation areas. As an application of SDINS/GPS integration, the in-flight alignment process of a SDINS utilizing GPS carrier phase measurements is introduced and analyzed via an observability analysis using nul1 space method. A measurement model of double-differenced GPS carrier phase measurements is newly derived in order to be used with a SDINS error model. Also, conditions for determining the complete observability of a SDINS/GPS system are suggested and proved. Consequently, it is shown that the system is not completely observable in case of one basel me. With one baseline aligned with y-axis of body frame, pitch error and x-axis accelerometer bias are unobservable states. Also shown is that al1 states are completely observable when sequential maneuver is performed. Above results are confirmed by a covariance analysis.

• Journal of the Korean Society for Railway
• /
• v.6 no.1
• /
• pp.15-20
• /
• 2003

The computer simulation techniques were adopted to evaluate the effects of seating positions of passenger under various accident scenarios. The baseline of computer simulation model was tuned by the sled impact tests which conducted under the upright and standard seating positions. This study shows the effect of relative velocity between occupant and struck vehicle while occupant is impacted to a front seat's seatback. Although, base on the current accident scenarios, The KHST is performed well enough to protect average adult male occupants. However, Results from the tests indicate small size occupant or higher impact speed may cause sever neck and femur injuries.

• International journal of advanced smart convergence
• /
• v.5 no.4
• /
• pp.1-9
• /
• 2016

In order to reconstruct a 3D model in video sequences, to select key frames that are easy to estimate a geometric model is essential. This paper proposes a method to easily extract informative frames from a handheld video. The method combines selection criteria based on appropriate-baseline determination between frames, frame jumping for fast searching in the video, geometric robust information criterion (GRIC) scores for the frame-to-frame homography and fundamental matrix, and blurry-frame removal. Through experiments with videos taken in indoor space, the proposed method shows creating a more robust 3D point cloud than existing methods, even in the presence of motion blur and degenerate motions.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.13 no.6
• /
• pp.24-33
• /
• 2009

The present study deals with the optimization of the leading edge shape around a wing-body junction to minimize the strength of the horseshoe vortex, which is one of the main factor generating the secondary flow losses. For this purpose, approximate optimization method is used for the optimization. The study is performed by using $FLUENT^{TM}$ and $iSIGHT^{TM}$. The total pressure coefficient for the optimized model was decreased about 9.79% compared with the baseline model.

• Journal of the military operations research society of Korea
• /
• v.28 no.1
• /
• pp.29-46
• /
• 2002

This study focuses primarily on the construction of the wartime sealift operation model from US to Korea. There are some uncertainties in the process of sealift operation such as the procurement rate of materiel in US, the distribution of KFS on four initial position locations at the start of the activation, and the number of ports and berths in the SPOES and SPODS. The sealift capability, based on the allocation of sealift assets such as the number of vessels, berths, and ports, is evaluated through simulation. The simulation is executed with a baseline wartime scenario and then the results are analyzed through a sensitivity analysis. The military planner may use of this model as a standard for establishing effective and concrete sealift operation plan in the near future.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2012.05a
• /
• pp.276-276
• /
• 2012

일반적으로 기후변화영향에 관한 연구수행을 위해 전지구기후모형(GCM; Global Climate Model)이 사용되고 있다. 하지만 GCM은 공간해상도(Spatial resolution)가 거칠기 때문에 수문학 분야에서 주로 사용되는 유역규모의 지역적인 스케일특성과 물리적 특징을 표현하는데 한계가 있다. 또한 GCM 기후변수들 중 강수량의 경우 한반도 지역의 6월과 10월 사이에 연강수량의 67% 이상이 집중되는 계절성을 반영하지 못하고 있으며, 높은 불확실성을 보이고 있다. 본 연구에서는 GCM 기반의 다지점 인공신경망기법을 적용한 상세화(Downscaling)를 실시하였다. GCM의 24개 2D변수에 대한 주성분분석을 실시하여 신경망의 학습인자로 사용하였으며, 학습, 검증 및 예측기간은 각각 1981~1995년, 1996~2000년, 2011~2100년으로 A1B 시나리오를 대상으로 상세화를 실시하였다. 또한, 여름철 태풍사상을 모의하기 위한 Stochastic Typhoon Simulation기법과 Baseline과 Projection 사이의 강수량 보정을 위한 Dynamic Quantile Mapping 기법을 적용하여, 강수량의 불확실성을 최소화 하고자 하였다.

• Current Research on Agriculture and Life Sciences
• /
• v.30 no.2
• /
• pp.97-104
• /
• 2012

Over the last century, drainage systems have become an integral component of agriculture. Climate observations and experiments using General circulation models suggest an intensification of the hydrologic cycle due to climate change. This study presents hydrologic simulations assessing the potential impact of climate change on subsurface drainage in Daegu, Republic of Korea. Historical and Long Ashton Research Station weather generator perturbed future climate data from 15 general circulation models for a field in Daegu were ran into a water management simulation model, DRAINMOD. The trends and variability in rainfall and Soil Excess Water ($SEW_{30}$) were assessed from 1960 to 2100. Rainfall amount and intensity were predicted to increase in the future. The predicted annual subsurface drainage flow varied from -35 to 40 % of the baseline value while the $SEW_{30}$ varied from -50 to 100%. The expected increases in subsurface drainage outflow require that more attention be given to soil and water conservation practices.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.3 no.3
• /
• pp.178-186
• /
• 2022

In order to support biomimicry technology development, it is necessary to develop an omnidirectional service platform which can recommend principles of biomimicry and business ideas, providing experts' networks and carrying out their relevant education and promotion on the ground of baseline data and application research materials related to biomimicry. This study was conducted to establish any probable plans for construction and utilization of the future open-platform which will collect and serve the technology of biomimicry. Accordingly, biological and ecological information databases were examined along with the appreciation of construction and management of major biomimicry DB, and, based on the materials from the interview of related experts, a customer journey map was schematized. Lastly, in order to suggest a mid-to-long-term target-model, the roles of a future biomimicry knowledge service-platform were determined along with the potential plans for its construction and management based on case analysis and customers' needs.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.344-358
• /
• 2022

Text summarization is the task of producing a shorter version of a long document while accurately preserving the main contents of the original text. Abstractive summarization generates novel words and phrases using a language generation method through text transformation and prior-embedded word information. However, newly coined words or out-of-vocabulary words decrease the performance of automatic summarization because they are not pre-trained in the machine learning process. In this study, we demonstrated an improvement in summarization quality through the contextualized embedding of BERT with out-of-vocabulary masking. In addition, explicitly providing precise pointing and an optional copy instruction along with BERT embedding, we achieved an increased accuracy than the baseline model. The recall-based word-generation metric ROUGE-1 score was 55.11 and the word-order-based ROUGE-L score was 39.65.