• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1491-1495
• /
• 2023

This study utilized automated machine learning (AutoML) to calculate Arctic ice surface temperature (IST). AutoML-derived IST exhibited a strong correlation coefficient (R) of 0.97 and a root mean squared error (RMSE) of 2.51K. Comparative analysis with deep neural network (DNN) models revealed that AutoML IST demonstrated good accuracy, particularly when compared to Moderate Resolution Imaging Spectroradiometer (MODIS) IST and ice mass balance (IMB) buoy IST. These findings underscore the effectiveness of AutoML in enhancing IST estimation accuracy under challenging polar conditions.

• Geomechanics and Engineering
• /
• v.17 no.6
• /
• pp.565-572
• /
• 2019

Spatial variability is an inherent characteristic of soil, and auto-correlation length (ACL) is a very important parameter in the reliability or probabilistic analyses of geotechnical engineering that consider the spatial variability of soils. Current methods for estimating the ACL need a large amount of laboratory or in-situ experiments, which is a great obstacle to the application of random field theory to geotechnical reliability analysis and design. To estimate the ACL reasonably and efficiently, we propose a micro-structure based numerical simulation method. The quartet structure generation set algorithm is used to generate stochastic numerical micro-structure of soils, and scanning electron microscope test of soil samples combined with digital image processing technique is adopted to obtain parameters needed in the QSGS algorithm. Then, 2-point correlation function is adopted to calculate the ACL based on the generated numerical micro-structure of soils. Results of a case study shows that the ACL can be estimated efficiently using the proposed method. Sensitivity analysis demonstrates that the ACL will become stable with the increase of mesh density and model size. A model size of $300{\times}300$ with a grid size of $1{\times}1$ is suitable for the calculation of the ACL of clayey soils.

• Journal of the Korean Statistical Society
• /
• v.29 no.2
• /
• pp.231-245
• /
• 2000

This paper considers a linear regression model with space and time data in where the disturbances follow spatially correlated error components. We provide the best linear unbiased predictor for the one way error components. We provide the best linear unbiased predictor for the one way error component model with spatial autocorrelation. Further, we derive two diagnostic test statistics for the assessment of model specification due to spatial dependence and random effects as an application of the Lagrange Multiplier principle.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 1993.10a
• /
• pp.207-215
• /
• 1993

With the aid of finite element method, this paper deals with the problem of structural response variability of transmission tower subjected to the spatial variability of material properties, Young's modulus herein. The spatial variability of material property are modeled as two-dimensional stochastic field which has an isotropic auto-correlation function. Response variability has been computed based on two numerical techniques, such as the Neumann expansion method in conjunction with the Monte Carlo simulation method. The results by these numerical methods are compared with those by the deterministic approach.

• Korean Journal of Remote Sensing
• /
• v.28 no.4
• /
• pp.357-367
• /
• 2012

This paper presents a two-stage methodology for anomaly detection from hyperspectral imagery that consists of transform-based feature extraction and selection, and computation of a local spatial auto-correlation statistic. First, principal component transform and 3D wavelet transform are applied to reduce redundant spectral information from hyperspectral imagery. Then feature selection based on global skewness and the portion of highly skewed sub-areas is followed to find optimal features for anomaly detection. Finally, a local indicator of spatial association (LISA) statistic is computed to account for both spectral and spatial information unlike traditional anomaly detection methodology based only on spectral information. An experiment using airborne CASI imagery is carried out to illustrate the applicability of the proposed anomaly detection methodology. From the experiments, anomaly detection based on the LISA statistic linked with the selection of optimal features outperformed both the traditional RX detector which uses only spectral information, and the case using major principal components with large eigen-values. The combination of low- and high-frequency components by 3D wavelet transform showed the best detection capability, compared with the case using optimal features selected from principal components.

• Survey Research
• /
• v.10 no.3
• /
• pp.1-19
• /
• 2009

This study investigates the utilities of spatial analysis in the context of survey research using Geographical Information System(GIS) and Geographically Weighted Regression (GWR) which take account of spatial heterogeneity. Many social phenomena involve spatial dimension, and with the development of GIS, GPS receiver, and online location-based services, spatial information can be collected and utilized more easily, and thus application of spatial analysis in the survey research is getting easier. The traditional OLS regression models which assume independence of observations and homoscedasticity of errors cannot handle spatial dependence problem. GWR is a spatial analysis technique which utilizes spatial information as well as attribute information, and estimated using geographically weighted function under the assumption that spatially close cases are more related than distant cases. Residential survey data from a Primary Autonomous District are used to estimate a model of public service satisfaction. The findings show that GWR handles the problem of spatial auto-correlation and increases goodness-of-fit of model. Visualization of spatial variance of effects of the independent variables using GIS allows us to investigate effects and relationships of those variables more closely and extensively. Furthermore, GIS and GWR analyses provide us a more effective way of identifying locations where the effect of variable is exceptionally low or high, and thus finding policy implications for social development.

• Communications for Statistical Applications and Methods
• /
• v.17 no.3
• /
• pp.413-421
• /
• 2010

In this study, we considered bivariate conditional auto-regressive model taking into account spatial association as well as correlation between the two dependent variables, which are the counts of murder and burglary. We conducted likelihood ratio test for checking over-dispersion issues prior to applying spatial poisson models. For the real application, we used the annual counts of violent crimes at 25 districts of Seoul in 2007. The statistical results are visually illustrated by geographical information system.

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.43 no.4 s.346
• /
• pp.31-38
• /
• 2006

In this paper, we propose an efficient preceding scheme for STBC-Spatial Muiltiplexing OFDM systems. In MIMO systems, the precoder is designed on the assumption that feedback channel information is perfectly known to transmitter and receiver. However, feedback delay and link errors in real environment make the transmitter use the incorrect channel information and consequently cause the performance degradation. The proposed precoder is designed to compensate for the performance degradation by the diversity gain provided by STBC. At the transmitter, the precoder for each subcarrier is constructed by using the index of codebook, subcarrier correlation, and auto correlation of channel. From the simulation results, STBC-spatial multiplexing OFDM outperforms the preceded-spatial multiplexing OFDM at $SER=10^{-3}$ when the Doppler frequency is greater than 60Hz.

• Korean Journal of Remote Sensing
• /
• v.39 no.6_1
• /
• pp.1341-1352
• /
• 2023

The normalized difference vegetation index (NDVI) derived from satellite images is a crucial tool to monitor forests and agriculture for broad areas because the periodic acquisition of the data is ensured. However, optical sensor-based vegetation indices(VI) are not accessible in some areas covered by clouds. This paper presented a synthetic aperture radar (SAR) based approach to retrieval of the optical sensor-based NDVI using machine learning. SAR system can observe the land surface day and night in all weather conditions. Radar vegetation indices (RVI) from the Sentinel-1 vertical-vertical (VV) and vertical-horizontal (VH) polarizations, surface elevation, and air temperature are used as the input features for an automated machine learning (AutoML) model to conduct the gap-filling of the Sentinel-2 NDVI. The mean bias error (MAE) was 7.214E-05, and the correlation coefficient (CC) was 0.878, demonstrating the feasibility of the proposed method. This approach can be applied to gap-free nationwide NDVI construction using Sentinel-1 and Sentinel-2 images for environmental monitoring and resource management.

• The Korean Journal of Applied Statistics
• /
• v.25 no.2
• /
• pp.293-300
• /
• 2012

When Binomial data are obtained, logit and logit mixed models are commonly used for small area estimation. Those models are known to have good statistical properties through the use of unit level information; however, data should be obtained as area level in order to use area level information such as spatial correlation or auto-correlation. In this research, we suggested a new small area estimator obtained through the combination of unit level information with area level information.