• Journal of IKEEE
• /
• v.18 no.4
• /
• pp.523-531
• /
• 2014

In this paper, a fast single image blind denoising algorithm is presented, where noise parameters are estimated by local statistics of an observed degraded image without a prior information about the additive noise. The estimated noise parameters are used to define the constraints on the noise detection which is coupled with the 1st-order Markov Random Field. In addition, an adaptive modified weighted Gaussian filter is introduced, where variable window sizes and weighting coefficients defined by the constraints are used to control the degree of the smoothness of the reconstructed image. The experimental results demonstrate the capability of the proposed algorithm. Please put the abstract of paper here.

• Journal of English Language & Literature
• /
• v.55 no.1
• /
• pp.129-152
• /
• 2009

The paper aims to prove symmetries of the left and right edges of a morphological unit through the analysis of English blends. Blends are the words derived from two existing words by snipping the middle parts out and putting the rest together, such as smog (

• Journal of Soil and Groundwater Environment
• /
• v.11 no.1
• /
• pp.37-44
• /
• 2006

This study summarizes advantages and disadvantages of numerical methods and compares ELLAM and LEZOOMPC to develop an efficient numerical modeling technique on contaminant transport. Eulerian-Lagrangian method and Eulerian method are commonly used numerical techniques. However Eulerian-Lagrangian method does not conserve mass globally and fails to treat boundary in a straightforward manner. Also, Eulerian method has restrictions on the size of Courant number and mesh Peclet number because of time truncation error. ELLAM (Eulerian Lagrangian Localized Adjoint Method) which has been popularly used for past 10 years in numerical modeling, is known for overcoming these numerical problems of Eulerian-Lagrangian method and Eulerian method. However, this study investigates advantages and disadvantages of ELLAM and suggests a change for the better. To figure out the disadvantages of ELLAM, the results of ELLAM, LEZOOMPC (Lagrangian-Eulerian ZOOMing Peak and valley Capturing), and visual MODFLOW are compared for four examples having different mesh Peclet numbers. The result of ELLAM generates numerical oscillation at infinite of mesh Peclet number, but that of LEZOOMPC yields accurate simulations. The simulation results suggest that the numerical error of ELLAM could be alleviated by adopting some schemes in LEZOOMPC. In other words, the numerical model which combines ELLAM with backward particle tracking, forward particle tracking, adaptively local zooming, and peak/valley capturing of LEZOOMPC can be developed for not only overcoming the numerical error of ELLAM, but also keeping the numerical advantage of ELLAM.