• Journal of Environmental Science International
• /
• v.28 no.4
• /
• pp.423-433
• /
• 2019

This study was carried out to identify the problems of the underground watersheds on Jeju Island, and to establish the hydraulic groundwater basin to be used as basis for the analysis of the groundwater model. In order to evaluate the adequacy of the groundwater basin on Jeju Island, a correlation analysis between elevation and groundwater level was conducted using data from 125 observation wells. The analysis, conducted with an elevation step of 100 m, exhibited values of R2 in the range 0.1653-0.8011. No clear correlation was observed between elevation and groundwater level. In particular, the eastern and western areas showed an inverse proportionality between elevation and groundwater level. The Kriging technique was used to analyze the underground water level data and to define the equipotential lines for all areas of Jeju Island. Eight groundwater watersheds were delineated by considering the direction of groundwater flow, the positions of the observation wells, and the long and short axes of the watersheds.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.28 no.1
• /
• pp.1-21
• /
• 2024

This review comprehensively explores the evolution, theoretical underpinnings, variations, and applications of diffusion models. Originating as a generative framework, diffusion models have rapidly ascended to the forefront of machine learning research, owing to their exceptional capability, stability, and versatility. We dissect the core principles driving diffusion processes, elucidating their mathematical foundations and the mechanisms by which they iteratively refine noise into structured data. We highlight pivotal advancements and the integration of auxiliary techniques that have significantly enhanced their efficiency and stability. Variants such as bridges that broaden the applicability of diffusion models to wider domains are introduced. We put special emphasis on the ability of diffusion models as a crucial foundation model, with modalities ranging from image, 3D assets, and video. The role of diffusion models as a general foundation model leads to its versatility in many of the downstream tasks such as solving inverse problems and image editing. Through this review, we aim to provide a thorough and accessible compendium for both newcomers and seasoned researchers in the field.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.42 no.5 s.305
• /
• pp.19-26
• /
• 2005

In this paper, a new method for obtaining three-dimensional shape of an object by measuring relative blur between images using wavelet analysis has been described. Most of the previous methods use inverse filtering to determine the measure of defocus. These methods suffer from some fundamental problems like inaccuracies in finding the frequency domain representation, windowing effects, and border effects. Besides these deficiencies, a filter, such as Laplacian of Gaussian, that produces an aggregate estimate of defocus for an unknown texture, can not lead to accurate depth estimates because of the non-stationary nature of images. We propose a new depth from defocus (DFD) method using wavelet analysis that is capable of performing both the local analysis and the windowing technique with variable-sized regions for non-stationary images with complex textural properties. We show that normalized image ratio of wavelet power by Parseval's theorem is closely related to blur parameter and depth. Experimental results have been presented demonstrating that our DFD method is faster in speed and gives more precise shape estimates than previous DFD techniques for both synthetic and real scenes.

• Journal of the Korean Magnetics Society
• /
• v.20 no.3
• /
• pp.100-105
• /
• 2010

For predicting magnetic signals due to the remanent magnetization distributed on a ferromagnetic ship hull, this paper presents an efficient methodology for solving inverse problems, where the material sensitivity analysis based on the continuum mechanics is combined with the equivalent magnetic models. To achieve this, the 3D magnetic charge model and the magnetic dipole moment model are introduced and material sensitivity formulae applicable to each equivalent model are derived. The formulae offer the first-order gradient information of an objective function with respect to the variation of the magnetic charge or magnetic dipole and so an optimal solution can be easily obtained regardless of the number of design variables. To validate the proposed method, the numerical results are comparison with the real measurements of a mock-up model.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.6A
• /
• pp.794-801
• /
• 2000

In this paper, the efficient new performance evaluation method for digital communication channels is suggested and verified its efficiency in terms of simulation run-tim for the digital satellite broadcasting satellite TV channel. In order to solve the difficulties of the existing Importance Sampling(IS) Technics, we adopted the discrete probability mass function(PMF) in the new method for estimating the statistical characteristics of received signals from the measured Nth order central moments. From the discrete probability mass function obtained with less number of the received signal than the one required in the IS technic, continuous cumulative probability function and its inverse function are exactly estimated by using interpolation and extrapolation technic. And the overall channel is simplified with encoding block, inner channel performance degra-dation modeing block which is modeled with the Uniform Random Number Generator (URNG) and concatenated Inverse Cummulative Pr bility Distribution function, and decoding block. With the simplified channel model, the overall performance evaluation can be done within a drastically reduced time. The simulation results applied to the nonlinear digital satellite broadcasting TV channel showed the great efficiency of the alogrithm in the sense of computer run time, and demonstrated that the existing problems of IS for the nonlinear satellite channels with coding and M-dimensional memory can be completely solved.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.17 no.6
• /
• pp.1529-1540
• /
• 1993

The quenching of steels by water is one of the imprtant problems in the applications of heat treatment, but the fundamental researches by way of theoretical approaches have not been satisfactorily improbed yet. This study aimed at measuring the exact subcooled transient boiling curve for cylindrical specimens and at conducting the analytical researches into the prediction of cooling curves, including the latent heat of phase transformation of steel. Experiments of quenching were made with cylindrical specimens of carbon steel S45C of diameters from 12 to 30 mm and with Cu specimens of 12 mm diameter respectively. The internal temperature of specimens during the quenching process was measured by C-A sheathed thermocouple. The heat fluxes were numerically calculated by the numerical method of inverse heat conduction problem, using the measured inner temperature of specimen as a boundary condition. In case of ${\Delta}T_{sub}=80K$, $q_{s}$ is as follows $q_{s}=2.02{\times}10^{5}{\Delta}T_{set}^{0.05}:{\Delta}T_{set}\le500K.$ And the numerical analysis of unsteady heat conduction during the quenching process was made and the cooling curves were calculated by TDMA method.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.17 no.3
• /
• pp.149-157
• /
• 2005

Typical water quality (WQ) parameters defined in the governing equation of the WQ model are the pollutant loads from atmosphere and watersheds, pollutant release rates from sediment, diffusion coefficient and reaction coefficient etc. The direct measurement of these parameters is very difficult as well as requires high cost. In this study, the pollutant budget equation including these parameters was used to construct the linear simultaneous equations. Based on these equations, the inverse problems were constructed and WQ parameter estimation method minimizing the sum of squared errors between the computed and observed amounts of the mass changes was suggested. WQ parameters, i.e., the atmospheric pollutant loads, sediment release rates, diffusion coefficients and reaction coefficient, were estimated using .this method by utilizing the vertical concentration profile data which has been observed in Cheonsu Bay and Ulsan Port. Values of the estimated parameters show a large temporal variation. However, this technique is persuasive in that the RHS (root mean square) error was less than $5.0\%$ of the observed value ranges and the agreement index was greater than 0.95.

• Journal of the Korea Computer Graphics Society
• /
• v.20 no.1
• /
• pp.1-11
• /
• 2014

Patients with cerebral palsy or arthritis have deformities in lower limb which cause unstable gait or posture and pains. Surgeons perform a deformity correction osteotomy with surgical plan. But sometimes they find the unexpected angular or rotational deformation after surgery. The problems are that there is no method to predict the result of a surgical plan and also there are so many factors to must consider in surgical planning step such as clinical measurements, rotation angle, wedge angle, morphology of lower limb, etc. This paper presents new methods for planning the deformity correction osteotomy efficiently. There are two approaches based on the 3D mesh model and the accurate assessment of the patient's lower limb. One is the manual pre-simulation of surgery using forward kinematics. And the other is the automatic surgical planning using inverse kinematics and nonlinear optimization. Using these methods, we can predict and verify the results of various surgical treatments and also we can find a more effective surgical plan easily compared to conventional methods.

• International Journal of Highway Engineering
• /
• v.10 no.4
• /
• pp.9-18
• /
• 2008

Curling is caused by the difference in the temperature and humidity by the depth of the slab in Jointed Plain Concrete Pavement. Slab curvature shape and size change due to curling exert a profound influence on the internal stress and roughness of the pavement, affecting structural and functional performance of the pavement. Direct measurement of the slab curvature entails many problems. Many measuring instruments have to be installed at the early-stage of the pavement construction, and the behavior of the slab curvature needs to be measured accurately from the early-stage. Moreover, the cost and technical difficulty are very formidable to measure the slab curvature. This study develops a measurement method for slab curvature in jointed concrete pavement at any given time by applying Power Spectrum Density Analysis and Inverse Fast Fourier Transformation to the profile data, that can be easily obtained at the construction field site. The effectiveness of this developed method is verified by measuring the profile data of the test road of jointed concrete pavement at an inland central expressway by the hour and by examining the result of extracting the slab curvature shape from this profile data. Additionally, the profile data of CRCP(Continuously Reinforced Concrete Pavement) sections on the same expressway were obtained and analyzed at the same time. The validity of the method developed for the slab curvature shape extraction is verified by comparing the result from the analysis of the profile data of CRCP sections with that from the analysis of the prof1Ie data of jointed concrete pavement sections.

• Geophysics and Geophysical Exploration
• /
• v.7 no.3
• /
• pp.207-212
• /
• 2004

This article reviews recent developments in three-dimensional (3-D) magntotelluric (MT) imaging. The inversion of MT data is fundamentally ill-posed, and therefore the resultant solution is non-unique. A regularizing scheme must be involved to reduce the non-uniqueness while retaining certain a priori information in the solution. The standard approach to nonlinear inversion in geophysis has been the Gauss-Newton method, which solves a sequence of linearized inverse problems. When running to convergence, the algorithm minimizes an objective function over the space of models and in the sense produces an optimal solution of the inverse problem. The general usefulness of iterative, linearized inversion algorithms, however is greatly limited in 3-D MT applications by the requirement of computing the Jacobian(partial derivative, sensitivity) matrix of the forward problem. The difficulty may be relaxed using conjugate gradients(CG) methods. A linear CG technique is used to solve each step of Gauss-Newton iterations incompletely, while the method of nonlinear CG is applied directly to the minimization of the objective function. These CG techniques replace computation of jacobian matrix and solution of a large linear system with computations equivalent to only three forward problems per inversion iteration. Consequently, the algorithms are efficient in computational speed and memory requirement, making 3-D inversion feasible.