• 한국공간구조학회논문집
• /
• 제20권4호
• /
• pp.159-167
• /
• 2020

The resilience performance evaluation method of a structure can evaluate the ability to recover after an earthquake disaster, and this study deals with the consideration and introduction of the resilience performance evaluation method. The resilience evaluation method can be expressed as a quantified number by constructing a loss estimation model and a recovery evaluation model. The recovery evaluation model should consider downtime in addition to the repair time, and the loss estimation model should consider not only direct loss to structures and non-structures, but also indirect loss due to functional loss of the building. In addition, to build a loss estimation model, the structure should be simplified to perform an efficient analysis. Therefore, in this study, the equivalent terminal induction system proposed cantilever-type and rahmen-type SDOF, and it is evaluated somewhat conservatively compared to the example structure, and it is judged that there is a need to improve the hysteresis characteristics by applying the stiffness reduction factor of the SDOF model.

• 한국통신학회논문지
• /
• 제42권2호
• /
• pp.323-330
• /
• 2017

밀리미터파 대역은 주파수가 30GHz-300GHz이고, 파장이 10mm-1mm인 EHF (Extremely High Frequency) 대역이다. 밀리미터파 대역에서는 장애물이 있는 경우 전파 감쇠가 심하기 때문에 Line-of-Sight (LoS)가 아닌 경우 신호가 잘 잡히지 않는다. 그렇기 때문에 밀리미터파 대역에서 신호 감쇠 측정을 할 경우에 측정 장비가 noise와 구별할 수 없는 신호들이 관찰된다. 이와 같이 감쇠가 심한 환경에서 신호 감쇠 data를 보면 특정한 값에서 제한을 받는 것이 관찰된다. 특정한 값에서 제한 받는 것을 그대로 두고 일반적인 Least square로 추정을 하는 경우에는 감쇠 exponent를 과소평가 할 수도 있다. 본 논문에서는 특정한 값에서 제한을 받아도 정확한 추정이 가능한 Tobit Maximum Likelihood Estimation, Heckman Two-stage Model 그리고 Truncation Regression model의 성능 비교를 하였다.

• 한국음향학회지
• /
• 제32권2호
• /
• pp.157-166
• /
• 2013

본 논문은 패킷 손실의 영향을 이용한 비 침입적 음질 평가 방법을 제안한다. 패킷 손실은 패킷 기반의 통신 시스템에서 음질을 저하시키는 주된 요소이며 그 영향은 코덱에 내장된 패킷 손실 은닉 알고리듬에 의해 결정된다. 패킷 손실 영향을 반영한 음질평가 시스템을 위해 VoIP 에서 협대역 코덱으로 사용되는 코덱 중 하나인 G.729를 선택하였으며, 음성 특징에 따른 패킷 손실 영향을 구분하기 위해서 G.729 코덱의 음성 파라미터를 이용한 한 음성 특성 분류기를 설계하였다. 이후, 각각의 패킷 특성에 따른 음질 저하의 정도를 수치화하기 위해 원 PESQ-LQ점수와 상관계수를 최대화하는 음질 저하 가중치를 반복적으로 구하였으며, 최종 음질 저하는 가중합으로 구하였다. 그 결과 제안한 모델과 PESQ-LQ의 상관계수는 칩입 모델 에서는 0.8950를, 비 침입 모델 에서는 0.8911의 결과를 나타내었다.

• Communications for Statistical Applications and Methods
• /
• 제2권2호
• /
• pp.52-62
• /
• 1995

Consider an estimation problem under squared error loss in a family of non-regular densities with both terminals of the support being decreasing functions of an unknown parameter. Using Karlin's(1958) technique, sufficient conditions are given for generalized Bayes estimators to be admissible for estimating an arbitrarily positive, monotone parametric function and then treat some examples which illustrate our results.

• Journal of the Korean Statistical Society
• /
• 제23권1호
• /
• pp.151-166
• /
• 1994

Let $X_1, \cdot, X_p$ be p independent random variables, where each $X_i$ has a distribution belonging to one parameter discrete power series distribution. The problem is to simultaneously estimate the unknown parameters under an asymmetric loss. Several new classes of dominating estimators are obtained by solving certain difference inequality.

• 대한임베디드공학회논문지
• /
• 제17권4호
• /
• pp.209-215
• /
• 2022

This paper proposes a novel approach for training an unsupervised depth estimation algorithm. The objective of unsupervised depth estimation is to estimate pixel-wise distances from camera without external supervision. While most previous works focus on model architectures, loss functions, and masking methods for considering dynamic objects, this paper focuses on the training framework to effectively use depth cue. The main loss function of unsupervised depth estimation algorithms is known as the photometric error. In this paper, we claim that direct depth cue is more effective than the photometric error. To obtain the direct depth cue, we adopt the technique of knowledge distillation which is a teacher-student learning framework. We train a teacher network based on a previous unsupervised method, and its depth predictions are utilized as pseudo labels. The pseudo labels are employed to train a student network. In experiments, our proposed algorithm shows a comparable performance with the state-of-the-art algorithm, and we demonstrate that our teacher-student framework is effective in the problem of unsupervised depth estimation.

• 한국건축시공학회지
• /
• 제23권4호
• /
• pp.485-496
• /
• 2023

Precise construction cost estimation is paramount to determining the total construction expense of a project prior to the initiation of the construction phase. Despite this, manual quantification and cost estimation methods, which continue to be widely used, may result in imprecise estimation and subsequent financial loss. Given the fast-paced and efficiency-demanding nature of the construction industry, trustworthy quantity and cost estimation is essential. To mitigate these obstacles, this research is focused on establishing an automated quantity estimation algorithm, particularly designed for the main rebar of beams which are recognized for their complicated reinforcement configurations. The exact quantity derived from the proposed algorithm is compared to the manually approximated quantity, reflecting a variance of 10.27%. As a result, significant errors and impending financial loss can be averted. The implementation of the findings from this research holds the potential to significantly assist construction firms in quickly and accurately estimating rebar quantities while adhering strictly to applicable specifications and regulatory requirements.

• 한국농공학회논문집
• /
• 제56권6호
• /
• pp.159-167
• /
• 2014

In this study the process of normalizing hydraulic head loss rate was developed for the purpose of estimation of seepage blocking state at each seepage segment in sea dike embankment. Pore water pressure sensors were installed with some interval along seepage path, then the hydraulic head loss rate at each segment between pore water pressure sensors was calculated, and then the calculated hydraulic head loss rate was normalized based on seepage path length. The comparison of normalized hydraulic head loss rates showed that the cross section of sea dike embankment was homogeneous approximately and the width of cross section was long enough to blocking tide water.

• 한국지하수토양환경학회지:지하수토양환경
• /
• 제19권6호
• /
• pp.30-36
• /
• 2014

The objective of this study is to estimate soil loss in the buffer zone of Guem river with future climate change simulation. Revised Universal Soil Loss Equation (RUSLE) model was used for the estimation of soil loss at the buffer zone of Guem river. As results of simulations, the area of the maximum soil loss potential was estimated as the Cheongsung-myeon Okchun-gun Chungcheongbuk-do. The soil losses were estimated to be 106.67 and 103.00 ton/ha/yr for the 2020 segi (2015-2025) and 2040 segi (2035-2045) in the Cheongsung-myeon area, respectively. Also, the estimated average values of soil losses in the Cheongsung-myeon with future climate change was 110.78 ton/ha/yr.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2009년도 춘계학술대회
• /
• pp.657-660
• /
• 2009

영상 보간법은 다양한 영상 처리를 위하여 사용되는 기반 기술로서, 보간 과정에서 발생하는 화질열화를 최소화하기 위한 연구가 활발히 진행되고 있다. 본 논문에서는 손실 정보 추정을 이용하여 개선된 양선형 보간법을 제안한다. 제안하는 방법에서는 획득된 저해상도 영상의 다운 샘플링 및 보간을 통하여 저해상도 영상 생성시 발생하는 손실 정보를 추정하고, 추정한 손실 정보를 고해상도로 보간된 영상에 적용하여 화질 열화를 최소화한다. 동일한 영상을 이용한 실험을 통해서 기존 방법들 보다 0.97~1.79dB의 PSNR이 향상된 것을 알 수 있었고, 윤곽선을 비롯한 주관적 화질 향상을 역시 확인하였다. 제안하는 방법은 영상 해상도 개선과 영상 복원을 위한 다양한 응용 환경에서 유용하게 사용될 수 있다.