• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.3
• /
• pp.629-635
• /
• 2017

In this paper, we proposed a nonlinear synthesis filter for noise reduction to reduce the effects of Gaussian noise and impulse noise. When the centralization of the local mask is judged to be Gaussian noise by the noise judgment, the weight value of the weight filter are applied differently according to the spatial weight filter and the pixel change by using the sample variance in the local mask. And if it is determined as the impulse noise, we proposed an algorithm that applies different weights of local histogram weight filter and standard median filter according to noise density of mask. In order to evaluate the performance of the proposed filter algorithm, we used PSNR(peak signal to noise ratio) and compared existing methods and proposed filter algorithm in the mixed noise environment with Gaussian noise, impulsive noise, and two noises mixed.

• Journal of Digital Convergence
• /
• v.11 no.11
• /
• pp.409-414
• /
• 2013

Naive Bayesian classifiers are a powerful and well-known type of classifiers that can be easily induced from a dataset of sample cases. However, the strong conditional independence assumptions can sometimes lead to weak classification performance. Normally, naive Bayesian classifiers use Gaussian distributions to handle continuous attributes and to represent the likelihood of the features conditioned on the classes. The probability density of attributes, however, is not always well fitted by a Gaussian distribution. Another eminent type of classifier is the neuro-fuzzy classifier, which can learn fuzzy rules and fuzzy sets using supervised learning. Since there are specific structural similarities between a neuro-fuzzy classifier and a naive Bayesian classifier, the purpose of this study is to apply learning distribution graphs constructed by a neuro-fuzzy network to naive Bayesian classifiers. We compare the Gaussian distribution graphs with the fuzzy distribution graphs for the naive Bayesian classifier. We applied these two types of distribution graphs to classify leukemia and colon DNA microarray data sets. The results demonstrate that a naive Bayesian classifier with fuzzy distribution graphs is more reliable than that with Gaussian distribution graphs.

• Proceedings of the KIEE Conference
• /
• 1996.07a
• /
• pp.56-58
• /
• 1996

The Preisach model needs a density function to simulate the hysteresis phenomena. To obtain this function, many experimental data obtained from the first order transition curves are required to get accurate density function. However, it is difficult to perform this procedure, especially for the hard magnetic materials. In this paper, we compare the density function obtained from the experimental data with that computed from the mathematical function like the Gaussian function, and propose a simple technique to get mathematical equation of the density function or Everett function which is obtained from the initial curve, major and minor loop.

• Journal of Communications and Networks
• /
• v.16 no.6
• /
• pp.600-612
• /
• 2014

In this paper, we consider the problem of spectrum sensing for advanced television systems committee (ATSC) digital television (DTV) signal detection. To exploit the cyclostationarity of the ATSC DTV signals, we employ spectral correlation density (SCD) as the decision statistic and propose an optimal detection algorithm. The major difficulty is in obtaining the probability distribution functions of the SCD. To overcome the difficulty, we probabilistically model the pilot frequency location and employ Gaussian approximation for the SCD distribution. Then, we obtain a practically implementable detection algorithm that outperforms the industry leading systems by 2-3 dB. We also propose various techniques that greatly reduce the system complexity with performance degradation by only a few tenths of decibels. Finally, we show how robust the system is to the estimation errors of the noise power spectral density level and the probability distribution of the pilot frequency location.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2011.08a
• /
• pp.144-144
• /
• 2011

Diagnostics of plasma density and temperature play an important role for monitoring plasma processing and Laser Thomson scattering is a one of the most accurate diagnostic technique for measuring plasma density and temperature because of none-perturbation to plasma among various diagnostic techniques invented to measure plasma density and temperature. I will briefly review Laser Thomson scattering experiment performed in KRISS and difficulties for measuring the electron velocity distribution such as Gaussian due to low signal-to-noise ratio with showing results that we got until now. This work is an intermediate step in a process that we will get a reliable data which shows physical phenomenon of plasma compared with other diagnostic techniques and results.

• Journal of Communications and Networks
• /
• v.11 no.5
• /
• pp.455-463
• /
• 2009

We consider the challenges of finding good puncturing patterns for rate-compatible low-density parity-check code (LDPC) codes over additive white Gaussian noise (AWGN) channels. Puncturing is a scheme to obtain a series of higher rate codes from a lower rate mother code. It is widely used in channel coding but it causes performance is lost compared to non-punctured LDPC codes at the same rate. Previous work, considered the role of survived check nodes in puncturing patterns. Limitations, such as single survived check node assumption and simulation-based verification, were examined. This paper analyzes the performance according to the role of multiple survived check nodes and multiple dead check nodes. Based on these analyses, we propose new algorithm to find a good puncturing pattern for LDPC codes over AWGN channels.

• Journal of Communications and Networks
• /
• v.14 no.5
• /
• pp.540-545
• /
• 2012

In this paper, the lagged cross-correlation of two probability density functions constructed by kernel density estimation is proposed, and by maximizing the proposed function, adaptive filtering algorithms for supervised and unsupervised training are also introduced. From the results of simulation for blind equalization applications in multipath channels with impulsive and slowly varying direct current (DC) bias noise, it is observed that Gaussian kernel of the proposed algorithm cuts out the large errors due to impulsive noise, and the output affected by the DC bias noise can be effectively controlled by the lag ${\tau}$ intrinsically embedded in the proposed function.

• Journal of The Korean Astronomical Society
• /
• v.46 no.3
• /
• pp.125-131
• /
• 2013

We present the relation between the genus in cosmology and the Betti numbers for excursion sets of three- and two-dimensional smooth Gaussian random fields, and numerically investigate the Betti numbers as a function of threshold level. Betti numbers are topological invariants of figures that can be used to distinguish topological spaces. In the case of the excursion sets of a three-dimensional field there are three possibly non-zero Betti numbers; ${\beta}_0$ is the number of connected regions, ${\beta}_1$ is the number of circular holes (i.e., complement of solid tori), and ${\beta}_2$ is the number of three-dimensional voids (i.e., complement of three-dimensional excursion regions). Their sum with alternating signs is the genus of the surface of excursion regions. It is found that each Betti number has a dominant contribution to the genus in a specific threshold range. ${\beta}_0$ dominates the high-threshold part of the genus curve measuring the abundance of high density regions (clusters). ${\beta}_1$ dominates the genus near the median thresholds which measures the topology of negatively curved iso-density surfaces, and ${\beta}_2$ corresponds to the low-threshold part measuring the void abundance. We average the Betti number curves (the Betti numbers as a function of the threshold level) over many realizations of Gaussian fields and find that both the amplitude and shape of the Betti number curves depend on the slope of the power spectrum n in such a way that their shape becomes broader and their amplitude drops less steeply than the genus as n decreases. This behaviour contrasts with the fact that the shape of the genus curve is fixed for all Gaussian fields regardless of the power spectrum. Even though the Gaussian Betti number curves should be calculated for each given power spectrum, we propose to use the Betti numbers for better specification of the topology of large scale structures in the universe.

• The Korean Journal of Cytopathology
• /
• v.7 no.1
• /
• pp.23-30
• /
• 1996

Chromatin texture, which partly reflects nuclear organization, is evolving as an important parameter indicating cell activation or transformation. In this study, chromatin pattern was evaluated by image analysis of the electron micrographs of follicular and papillary carcinoma cells of the thyroid gland and tested for discrimination of the two neoplasms. Digital grey images were converted from the electron micrographs, nuclear images, excluding nucleolus and intranuclear cytoplasmic inclusions, were obtained by segmentation; grey levels were standardized; and grey level histograms were generated. The histograms in follicular carcinoma showed Gaussian or near-Gaussian distribution and had a single peak, whereas those in papillary carcinoma had two peaks(bimodal), one at the black zone and the other at the white zone. In papillary carcinoma, the peak in the black zone represented an increased amount of heterochromatin particles and that at the white zone represented decreased electron density of euchromatin or nuclear matrix. These results indicate that the nuclei of follicular and papillary carcinoma cells differ in their chromatin pattern and the difference may be due to decondensed chromatin and/or matrix substances.

• Journal of Electrical Engineering and Technology
• /
• v.9 no.6
• /
• pp.2118-2125
• /
• 2014

Gait Recognition is a new technique to identify the people by the way they walk. Human gait is a spatio-temporal phenomenon that typifies the motion characteristics of an individual. The proposed method makes a simple but efficient attempt to gait recognition. For each video file, spatial silhouettes of a walker are extracted by an improved background subtraction procedure using Gaussian Mixture Model (GMM). Here GMM is used as a parametric probability density function represented as a weighted sum of Gaussian component densities. Then, the relevant features are extracted from the silhouette tracked from the given video file using the Principal Component Analysis (PCA) method. The Fisher Linear Discriminant Analysis (FLDA) classifier is used in the classification of dimensional reduced image derived by the PCA method for gait recognition. Although gait images can be easily acquired, the gait recognition is affected by clothes, shoes, carrying status and specific physical condition of an individual. To overcome this problem, it is combined with footprint as a multimodal biometric system. The minutiae is extracted from the footprint and then fused with silhouette image using the Discrete Stationary Wavelet Transform (DSWT). The experimental result shows that the efficiency of proposed fusion algorithm works well and attains better result while comparing with other fusion schemes.