• Korean Journal of Remote Sensing
• /
• v.35 no.6_3
• /
• pp.1273-1283
• /
• 2019

This paper proposes an advanced crop classification model that combines a procedure for weighted combination of spatial features extracted from multi-scale input images with a conventional convolutional neural network (CNN) structure. The proposed model first extracts spatial features from patches with different sizes in convolution layers, and then assigns different weights to the extracted spatial features by considering feature-specific importance using squeeze-and-excitation block sets. The novelty of the model lies in its ability to extract spatial features useful for classification and account for their relative importance. A case study of crop classification with multi-temporal Landsat-8 OLI images in Illinois, USA was carried out to evaluate the classification performance of the proposed model. The impact of patch sizes on crop classification was first assessed in a single-patch model to find useful patch sizes. The classification performance of the proposed model was then compared with those of conventional two CNN models including the single-patch model and a multi-patch model without considering feature-specific weights. From the results of comparison experiments, the proposed model could alleviate misclassification patterns by considering the spatial characteristics of different crops in the study area, achieving the best classification accuracy compared to the other models. Based on the case study results, the proposed model, which can account for the relative importance of spatial features, would be effectively applied to classification of objects with different spatial characteristics, as well as crops.

• Steel and Composite Structures
• /
• v.37 no.6
• /
• pp.695-709
• /
• 2020

The buckling properties of a single-layered graphene sheet (SLGS) are examined using nonlocal integral first shear deformation theory (FSDT) by incorporating the influence of visco-Pasternak's medium. This model contains only four variables, which is even less than the conventional FSDT. The visco-Pasternak's medium is introduced by considering the damping influence to the conventional foundation model which modeled by the linear Winkler's coefficient and Pasternak's (shear) foundation coefficient. The nanoplate under consideration is subjected to compressive in- plane edge loads per unit length. The impacts of many parameters such as scale parameter, aspect ratio, the visco-Pasternak's coefficients, damping parameter, and mode numbers on the stability investigation of the SLGSs are examined in detail. The obtained results are compared with the corresponding available in the literature.

• Korean Journal of Child Studies
• /
• v.33 no.6
• /
• pp.85-107
• /
• 2012

In this study, the Infants & Toddlers' Playfulness Scale(ITPS) was developed and its reliability and validity were examined. The participants in this study consisted of 656 parents whose children were 8 to 36 months old. Five factors were identified from the results of the exploratory factor analysis performed on the 29 items that had been derived by analyzing existing research. The Goodness of Fit Indices(GFIs) of the confirmatory factor analyses, performed on the 1st-order 5-factor model and the 2nd-order one factor model, were both satisfactory. Concurrent validity was established by the high correlations of r = .631~.808(p < .01) between each factor of the ITPS and the total score of the Children's Playfulness Scale(CPS), an existing measure for children's playfulness. The reliability of each factor, as measured by Cronbach's ${\alpha}$, ranged from .773 to .883. Consequently, the ITPS developed in this study can be seen as a reliable and valid scale that can be used by parents to measure the playfulness of their infants and toddlers in a multi-dimensional manner.

• Advances in nano research
• /
• v.4 no.4
• /
• pp.251-264
• /
• 2016

In this paper, a simple and refined nonlocal hyperbolic higher-order beam theory is proposed for bending and vibration response of nanoscale beams. The present formulation incorporates the nonlocal scale parameter which can capture the small scale effect, and it considers both shear deformation and thickness stretching effects by a hyperbolic variation of all displacements across the thickness without employing shear correction factor. The highlight of this formulation is that, in addition to modeling the displacement field with only two unknowns, the thickness stretching effect (${\varepsilon}_z{\neq}0$) is also included in the present model. By utilizing the Hamilton's principle and the nonlocal differential constitutive relations of Eringen, the equations of motion of the nanoscale beam are reformulated. Verification studies demonstrate that the developed theory is not only more accurate than the refined nonlocal beam theory, but also comparable with the higher-order shear deformation theories which contain more number of unknowns. The theoretical formulation proposed herein may serve as a reference for nonlocal theories as applied to the static and dynamic responses of complex-nanobeam-system such as complex carbon nanotube system.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.413-417
• /
• 2004

Recent experiments have shown the 'size effects' in micro/nano scale. But the classical plasticity theories can not predict these size dependent deformation behaviors because their constitutive models have no characteristic material length scale. The Mechanism - based Strain Gradient(MSG) plasticity is proposed to analyze the non-uniform deformation behavior in micro/nano scale. The MSG plasticity is a multi-scale analysis connecting macro-scale deformation of the Statistically Stored Dislocation(SSD) and Geometrically Necessary Dislocation(GND) to the meso-scale deformation using the strain gradient. In this research we present a study of nano-indentation by the MSG plasticity. Using W. D. Nix and H. Gao s model, the analytic solution(including depth dependence of hardness) is obtained for the nano indentation , and furthermore it validated by the experiments.

• Proceedings of the KSME Conference
• /
• 2007.05a
• /
• pp.527-531
• /
• 2007

The classical continuum theory-based thin film model is independent of their size and the surface effect can be ignored. But the surface to bulk ratio becomes very large in nano-size structures such as nano film, nano wire and nano beam. In this case, surface effect plays an important role and its contribution of the surface effect must be considered. Molecular dynamics simulation has been a conventional way to analyze these ultra-thin structures but structures in the range between submicro and micro are difficult to analyze by classical molecular dynamics due to the restriction of computing resources and time. Therefore, in present study, the continuum-based method is considered to predict the overall physical and mechanical properties of the structures in nano-scale, especially, for the thin-film. The proposed continuum based-thin plate finite element is efficient and reliable for the prediction of nano-scale film behavior.

• Nuclear Engineering and Technology
• /
• v.46 no.5
• /
• pp.655-666
• /
• 2014

The CUPID code has been developed at KAERI for a transient, three-dimensional analysis of a two-phase flow in light water nuclear reactor components. It can provide both a component-scale and a CFD-scale simulation by using a porous media or an open media model for a two-phase flow. In this paper, recent advances in the CUPID code are presented in three sections. First, the domain decomposition parallel method implemented in the CUPID code is described with the parallel efficiency test for multiple processors. Then, the coupling of CUPID-MARS via heat structure is introduced, where CUPID has been coupled with a system-scale thermal-hydraulics code, MARS, through the heat structure. The coupled code has been applied to a multi-scale thermal-hydraulic analysis of a pool mixing test. Finally, CUPID-SG is developed for analyzing two-phase flows in PWR steam generators. Physical models and validation results of CUPID-SG are discussed.

• Advances in nano research
• /
• v.7 no.6
• /
• pp.443-457
• /
• 2019

In this investigation, dynamic and bending behaviors of isolated protein microtubules are analyzed. Microtubules (MTs) can be considered as bio-composite structures that are elements of the cytoskeleton in eukaryotic cells and posses considerable roles in cellular activities. They have higher mechanical characteristics such as superior flexibility and stiffness. In the modeling purpose of microtubules according to a hollow beam element, a novel single variable sinusoidal beam model is proposed with the conjunction of modified strain gradient theory. The advantage of this model is found in its new displacement field involving only one unknown as the Euler-Bernoulli beam theory, which is even less than the Timoshenko beam theory. The equations of motion are constructed by considering Hamilton's principle. The obtained results are validated by comparing them with those given based on higher shear deformation beam theory containing a higher number of variables. A parametric investigation is established to examine the impacts of shear deformation, length scale coefficient, aspect ratio and shear modulus ratio on dynamic and bending behaviors of microtubules. It is remarked that when length scale coefficients are almost identical of the outer diameter of MTs, microstructure-dependent behavior becomes more important.

• Korean Journal of Remote Sensing
• /
• v.26 no.1
• /
• pp.29-38
• /
• 2010

It is necessary to develop an integration model which can account for various data acquired at different measurement scales in environmental thematic mapping with high-resolution ground survey data and relatively low-resolution remote sensing data. This paper presents and applies a multi-scale geostatistical methodology for downscaling of thematic maps generated from lowresolution remote sensing data. This methodology extends a traditional ordinary kriging system to a block kriging system which can account for both ground data and remote sensing data which can be regarded as point and block data, respectively. In addition, stochastic simulation based on block kriging is also applied to describe spatial uncertainty attached to the downscaling. Two downscaling experiments including SRTM DEM and MODIS Leaf Area Index (LAI) products were carried out to illustrate the applicability of the geostatistical methodology. Through the experiments, multi-scale geostatistics based on block kriging successfully generated relatively high-resolution thematic maps with reliable accuracy. Especially, it is expected that multiple realizations generated from simulation would be effectively used as input data for investigating the effects of uncertain input data on GIS model outputs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.12
• /
• pp.4420-4438
• /
• 2021

Image dehazing is an ill-posed problem which is far from being solved. Traditional image dehazing methods often yield mediocre effects and possess substandard processing speed, while modern deep learning methods perform best only in certain datasets. The haze removal effect when processed by said methods is unsatisfactory, meaning the generalization performance fails to meet the requirements. Concurrently, due to the limited processing speed, most dehazing algorithms cannot be employed in the industry. To alleviate said problems, a lightweight fast dehazing network based on a multiple scale-patch framework (MSP) is proposed in the present paper. Firstly, the multi-scale structure is employed as the backbone network and the multi-patch structure as the supplementary network. Dehazing through a single network causes problems, such as loss of object details and color in some image areas, the multi-patch structure was employed for MSP as an information supplement. In the algorithm image processing module, the image is segmented up and down for processed separately. Secondly, MSP generates a clear dehazing effect and significant robustness when targeting real-world homogeneous and nonhomogeneous hazy maps and different datasets. Compared with existing dehazing methods, MSP demonstrated a fast inference speed and the feasibility of real-time processing. The overall size and model parameters of the entire dehazing model are 20.75M and 6.8M, and the processing time for the single image is 0.026s. Experiments on NTIRE 2018 and NTIRE 2020 demonstrate that MSP can achieve superior performance among the state-of-the-art methods, such as PSNR, SSIM, LPIPS, and individual subjective evaluation.