• Journal of Information Processing Systems
• /
• v.19 no.3
• /
• pp.275-288
• /
• 2023

Many image edge details are easily lost in the image denoising process, and the smooth image regions are prone to produce jagged. In this paper, we propose a wavelet-based image global k- singular value decomposition variational method to remove image noise. A layer of wavelet decomposition is applied to the noisy image first. Then, the image global k-singular value decomposition (IGK-SVD) method is used to remove the random noise of low-frequency components. Furthermore, a constructed variational denoising method (VDM) removes the random noise in the high-frequency component. Finally, the denoised image is obtained by wavelet reconstruction. The experimental results show that the proposed method's peak signal-to-noise ratio (PSNR) value is higher than other methods, and its structural similarity (SSIM) value is closer to one, indicating that the proposed method can effectively suppress image noise while retaining more image edge details. The denoised image has better denoising effects.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.11-16
• /
• 2005

Recently, there are a lot of endeavor to implement a fuzzy theory to artificial neural network. Goh proposed the fuzzy single layer perceptron algorithm and advanced fuzzy perceptron based on the generalized delta rule to solve the XOR Problem and the classical Problem. However, it causes an increased amount of computation and some difficulties in application of the complicated image recognition. In this paper, we propose an enhanced fuzzy single layer Perceptron using the dynamic adjustment of threshold. This method is applied to the XOR problem, which used as the benchmark in the field of pattern recognition. The method is also applied to the recognition of digital image for image application. In a result of experiment, it does not always guarantee the convergence. However, the network show improved the learning time and has the high convergence rate.

• Aerospace Engineering and Technology
• /
• v.11 no.1
• /
• pp.118-125
• /
• 2012

The purpose of this study is to analyze the ortho image quality change according to the generation technique of Digital Elevation Model(DEM) based on the digital map. First of all, two different types of DEM were generated using contour layer(Case1), contour layer and altitude layer(Case2) from the digital map on the scale of 1/5,000. After generating and evaluating two types of DEM, KOMPSAT-2 ortho images were generated by using them. In conclusion, Case2 DEM was more effective to use in the slope and switchback area, on the other hand, Case1 DEM was much better than Case2 DEM for preventing a topographic distortion in flat area.

• Journal of the Korean Society of Visualization
• /
• v.8 no.1
• /
• pp.46-52
• /
• 2010

In the present study, we employed the confocal laser scanning microscopy (CLSM) system to visualize the blood flow field with $1{\times}1{\mu}m^2$ spatial resolution. Based on the confocal microscopic image of red blood cells (RBCs), we performed the velocity vector field measurement and evaluated characteristics of cell migration from the cell depleted layer thickness calculation. The rat and mouse's blood were supplied into a micro glass tubes in vitro. The line scanning rate of confocal microscopy was 15 kHz for a $500{\times}500$ pixels image. As a result, the red blood cell itself can be used as a tracer directly without any kind of invasive tracer particle to get the velocity vector field of blood flow by performing particle image velocimetry (PIV) technique.

• Proceedings of the KSME Conference
• /
• 2003.11a
• /
• pp.610-615
• /
• 2003

This experiment has been carried out to find the structure of turbulent boundary layer with instantaneous velocity fields obtained in stream-wall-normal planes using a stereo-PIV (Particle Image Velocimetry) method. And it has been measured perpendicular plane and horizontal plane with hairpin vortex structure by Reynolds number change and made third dimension shape for section of horizontal plane through stereo-PIV. In the outer layer hairpin vortices occur in streamwise-aligned packets that propagate with small velocity dispersion. A streaky structure is composed of counter-rotating vortex. According as y+ increases, streaky structure's interval space decrease, and it shows that hairpin shape of prior research is vertified. The objective of the present research is to gain a better understanding of coherent structures in the outer of wall turbulence by experimentally examining coherent structures.

• Journal of Advanced Marine Engineering and Technology
• /
• v.29 no.4
• /
• pp.453-460
• /
• 2005

Experiments were conducted to investigate the effect of the surface roughness on the flat plate turbulent boundary layer. The square rods were installed at the leading edge to make surface roughness. The particle image velocimetry was used to measure the mean velocities and velocity fluctuation component. All measurements were made over a range of w/k=1. 2 5 and $Re_x=80.000{\sim}360,000$. Friction velocity was measured by using Clauser plot method. The level of turbulent intensities on roughness surface appears more strongly than that of turbulent intensities on flat plate. A correlation of boundary layer thickness in term of $Re_x$ and w/k are presented.

• The Journal of the Korean dental association
• /
• v.57 no.1
• /
• pp.26-33
• /
• 2018

Panoramic radiography is widely used as a primary radiological diagnosis in dentistry. However, due to the overlapping of anatomical structures, diagnostic capability of panoramic radiography is limited. Recently, panoramic machine equipped with various image processing software has been introduced to improve the diagnostic capability. This study aims to introduce panoramic equipment with multi-layer imaging software that can provide 41 images from anterior to posterior in region of interest, and to report in various clinical applications.

• Transactions of the Korean hydrogen and new energy society
• /
• v.30 no.5
• /
• pp.455-464
• /
• 2019

Experiments were conducted to investigate the sensitivity of steel plate oxidation with temperature in a simulated furnace. Used steel plates were a general steel and a high tensile steel. Porous media burner (PM burner) used in model furnace was made for uniform temperature profile. The surrounding temperature was controlled by adjusting the flow rate of the mixture in the combustor. Oxide layer analysis was performed using SEM image analysis and EDS line scanning. Both steel sheets showed a tendency to increase the thickness of the steel sheet surface oxide layer as the temperature increases, and it was confirmed that the flaking phenomenon in surface oxidation layer appeared when the temperature was above a certain temperature.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.121-130
• /
• 2020

This paper suggests evaluating the military closed network data as an image which is generated by Generative Adversarial Network (GAN), applying an image evaluation method such as the InceptionV3 model-based Inception Score (IS) and Frechet Inception Distance (FID). We employed the famous image classification models instead of the InceptionV3, added layers to those models, and converted the network data to an image in diverse ways. Experimental results show that the Densenet121 model with one added Dense Layer achieves the best performance in data converted using the arctangent algorithm and 8 * 8 size of the image.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.11a
• /
• pp.799-802
• /
• 2017

We attempt to combines concepts of Morphological Operator(MO) and Convolutional Neural Networks(CNN) to improve image-to-image translation. To do this, we propose an operation that approximates morphological operations. Also we propose S-Convolution, an operation that extends the operation to use multiple filters like CNN. The experiment result shows that it can learn MO with big filter using multiple S-convolution layer of small filter. To validate effectiveness of the proposed layer in image-to-image translation we experiment with GAN with S-convolution applied. The result showed that GAN with S-convolution can achieve distinct result from that of GAN with CNN.