• Journal of Information Processing Systems
• /
• v.18 no.6
• /
• pp.794-802
• /
• 2022

The recovery of reasonable depth information from different scenes is a popular topic in the field of computer vision. For generating depth maps with better details, we present an efficacious monocular depth prediction framework with coordinate attention and feature fusion. Specifically, the proposed framework contains attention, multi-scale and feature fusion modules. The attention module improves features based on coordinate attention to enhance the predicted effect, whereas the multi-scale module integrates useful low- and high-level contextual features with higher resolution. Moreover, we developed a feature fusion module to combine the heterogeneous features to generate high-quality depth outputs. We also designed a hybrid loss function that measures prediction errors from the perspective of depth and scale-invariant gradients, which contribute to preserving rich details. We conducted the experiments on public RGBD datasets, and the evaluation results show that the proposed scheme can considerably enhance the accuracy of depth prediction, achieving 0.051 for log10 and 0.992 for δ<1.25³ on the NYUv2 dataset.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.269-273
• /
• 2008

기상청(KMA, Korea Meteorological Administration)에서는 기상수치예보모델을 적용하여 수치예보를 하고 있으며 전지구 모델로는 GDAPS(Global Date Assimilation and Prediction System)를 지역모델은 RDAPS(Regional Date Assimilation and Prediction System)를 사용하고 있다. 수치예보결과를 이용하여 유출량을 예측할 경우 일반적으로 해상도가 높은 지역모델인 RDAPS의 수치예보 결과값을 사용한다. RDAPS는 00UTC와 12UTC에 3시간으로 누적된 자료를 30km 격자에 대하여 예측시간으로부터 48시간에 대하여 자료를 생성한다. 일강우자료를 입력자료로 사용하는 강우-유출 모형의 경우 3시간 누적 자료를 나타나는 RDAPS 수치예보 결과를 이용 시 3시간 scale에서 일(day)시간 scale로 변환시켜주어야 한다. 본 연구에서는 RDAPS의 수치예보 결과의 일(day)시간 scale 변환 방법에 따른 정확도를 비교하여 RDAPS 수치예보 결과의 일(day)시간 scale 변환에 대한 정확도를 비교하여 일(day)시간 scale 변환에 대한 지침을 제공하고자 한다. RDAPS 수치예보 결과값의 특징을 이용하여 RDAPS 결과값을 일(day)시간 scale로 변환하는 방법으로 총 9개방법을 적용하였으며, 참 값으로는 기상청 강수자료를 사용하였으며, 금강유역을 대상으로 유역평균강수량을 계산하여 각 변환 방법에 따른 정확도를 비교하였다.

• Proceedings of the KSR Conference
• /
• 2011.10a
• /
• pp.2844-2851
• /
• 2011

The purpose of this study is to set up an acoustic prediction technique and to perform the IL test of scale down noise reducing device for the development of the noise reducing device as the development of 400km/h class high speed train. First of all, the IL prediction of noise reducing device was performed with the 2D BEM method. And the noise test of scale down noise reducing device in anechoic chamber was performed for the verification of acoustic prediction technique and IL performance evaluation. As the results, the acoustic prediction technique for the development of noise reducing device was verified because the averaged IL difference between prediction and test is in 2dB(A). And the measured IL value of noise reducing device is less than 2dB(A), and additional IL with polyester absorption material is increased about 0.5dB(A).

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.20-21
• /
• 2019

The Joint Video Experts Team (JVET) has been working on the development of next generation of video coding standard called Versatile Video Coding (VVC). Position Dependent Intra Prediction Combination (PDPC) which is one of the major tools for intra prediction refines the prediction through a linear combination between the reconstructed samples and the predicted samples according to the sample position. In VVC WD6, nScale which is shift value that adjusts the weight is determined by the width and height of the current block. It may cause that PDPC is applied to regions that do not fit the characteristics of the current intra prediction mode. In this paper, we define nScale for each width and height so that the weight can be applied independently to the left and top reference samples, respectively. Experimental results show that, compared to VTM 6.0, the proposed method gives -0.01%, -0.04% and 0.01% Bjotegaard-Delta (BD)-rate performance, for Y, Cb, and Cr components, respectively, in All-Intra (AI) configuration.

• Wind and Structures
• /
• v.36 no.6
• /
• pp.367-377
• /
• 2023

This study proposes a few-shot learning model for extrapolating the wind pressure of scaled experiments to full-scale measurements. The proposed ML model can use scaled experimental data and a few full-scale tests to accurately predict the remaining full-scale data points (for new specimens). This model focuses on extrapolating the prediction to different scales while existing approaches are not capable of accurately extrapolating from scaled data to full-scale data in the wind engineering domain. Also, the scaling issue observed in wind tunnel tests can be partially resolved via the proposed approach. The proposed model obtained a low mean-squared error and a high coefficient of determination for the mean and standard deviation wind pressure coefficients of the full-scale dataset. A parametric study is carried out to investigate the influence of the number of selected shots. This technique is the first of its kind as it is the first time an ML model has been used in the wind engineering field to deal with extrapolation in wind performance prediction. With the advantages of the few-shot learning model, physical wind tunnel experiments can be reduced to a great extent. The few-shot learning model yields a robust, efficient, and accurate alternative to extrapolating the prediction performance of structures from various model scales to full-scale.

• Proceedings of the KSME Conference
• /
• 2008.11b
• /
• pp.2548-2553
• /
• 2008

Prediction of the rotor blade performance is important for determining design factors such as weight and size in development of a small-scale helicopter. Generally, prediction of helicopter performance means the estimation of the power required for a given flight condition. However, due to lack of test data and analyzed results for small-scale rotor blade operated at low Reynolds numbers ($Re{\approx}10^5$), this is not an easy task. As an initial research, this work performs a modeling of a single rotor configuration with FLIGHTLAB and a experimental research with rotor test bed. In this process, we performed small-scale isolated single rotor by experimental and numerical method and achieved good agreement of the hover performance on the test data and simulation results.

• Proceedings of the KSR Conference
• /
• 2003.10c
• /
• pp.132-137
• /
• 2003

This paper concerns the prediction of railway noise propagation using scale model experiment in acoustics. In order to make acoustical experiment the digital signal processing technique are applied and spark discharge sound sources have been developed in which impulse response measured in 1/20 scale model railway. In the case of scale model experiment, it is difficult to realize sufficiently small size and directivity and to get sufficient sound energy and to get repeatability. Several type of Spark discharge sound source is made in laboratory. Experiment results are compared with the calculated results by the prediction model. As the results, it was found that railway noise could be predicted in acoustical scale model experiment using spark discharge sound source.

• 제어로봇시스템학회:학술대회논문집
• /
• 1986.10a
• /
• pp.547-549
• /
• 1986

This paper deals with large scale distributed packet switching system which is modeled by state space form and optimizing routing algorithms and buffer size via a hierachical system optimization method, the interaction prediction method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1987.10b
• /
• pp.223-227
• /
• 1987

This paper presents the extended interaction prediction method for large scale discrete-time systems with interconnected state and control. Feedback gain is obtained from decentralized calculation without solving Riccati equation. Hence, Computer storage and calculation time is reduced.

• Journal of East-West Nursing Research
• /
• v.12 no.1
• /
• pp.5-13
• /
• 2006

Purpose: This research has been conducted in order to confirm the major factors that prediction organizational commitment in registered nurses. Method: The subjects were 350 registered nurses from 3 hospitals in Seoul. The sample for data collection consisted of 329 useable questionnaires (94% overall return rate) for 2 weeks. The Instrument tools utilized in this study were organizational commitment scale, empowerment scale, job stress scale and job satisfaction scale and thoroughly modified to verify validity and reliability. The collected data have been analyzed using SPSS 11.0 program. Three outliers which were bigger than 3 in absolute value were found, so after taking them off, Multiple Regression was used for further analysis. Result: The major factors that prediction organizational commitment in registered nurses were job satisfaction, empowerment, age and unit experience, which explained 51.9% of organizational commitment. Conclusion: It has been confirmed that the regression equation model of this research may serve as a organizational commitment prediction factors in Registered Nurses.