• Proceedings of the KOSOMBE Conference
• /
• v.1996 no.11
• /
• pp.229-232
• /
• 1996

we propose the shrinkage of wavelet coefficients and the averaging method. The wavelet analysis decomposes the measured evoked potentials into scale coefficients and wavelet coefficients as a resolution level, respectively. And in the course of synthesis of evoked potentials, the presented method shrinks the wavelet coefficients, and then reproduces the evoked potentials and lastly averages it. we measured VEP signal to simulate the presented method, and compared it with averaged signal and LMS algorithm. As a result of simulations, the proposed method gets improved VEP about 0.2-1.6db in comparison with the result of averaging method.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.303-307
• /
• 2002

Although Synthetic Aperture Radar(SAR) is a very powerful and attractive tool, automatic interpretation of SAR images is extremely difficult because of several reason. Spatially, speckle noise reduction in SAR images is important step to interpret the SAR image at the preprocessing step. The speckle noise in SAR images is modeled to be multiplicative, and therefore, a signal-dependent noise. So, it has deflated many image-denoising algorithms that are based on additive noise model. In this paper, we propose an adaptive wavelet shrinkage method for speckle noise reduction in SAR images by analyzing the high frequency level in detail. We first decompose minutely the high frequency level to analyze the noise level. And then, we determine the weighting threshold value per the level, and layer. Finally, using those weighting threshold, we produce the efficient wavelet shrinkage method. So, this method not only reduces the speckle noise, but also preserves image detail and sharpness.

• Korean Journal of Mathematics
• /
• v.22 no.2
• /
• pp.289-305
• /
• 2014

In this paper we propose a new approach to the variable selection problem for a primary resolution and wavelet basis functions in wavelet regression. Most wavelet shrinkage methods focus on thresholding the wavelet coefficients, given a primary resolution which is usually determined by the sample size. However, both a primary resolution and the basis functions are affected by the shape of an unknown function rather than the sample size. Unlike existing methods, our method does not depend on the sample size and also takes into account the shape of the unknown function.

• Journal of the Institute of Convergence Signal Processing
• /
• v.15 no.4
• /
• pp.141-147
• /
• 2014

The wavelet shrinkage is a technique that reduces the wavelet coefficients to minimize the MSE(Mean Square Error) between the signal and the noisy signal by making use of the threshold determined by the variance of the wavelet coefficients. In this paper, by using the monotonic transform and the power of wavelet subbands, new thresholds applicable to the high and the low frequency wavelet bands are proposed, and the thresholds are applied to the ST(soft-threshold) technique to denoise on image signals with additive Gaussian noise. And the results of PSNRs are compared with the results obtained by the VisuShrink technique and those of [15]. The results shows the validity of this technique.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.521-524
• /
• 1999

In spite of advances in image resolution and film contrast, check screen/film mammography remains one of diagnostic imaging modality where the image interpretation is very difficult. For the enhancement of film mammography, in this paper, dyadic wavelet transform is introduced. An unsharp masking technique is proposed and performed in wavelet domain. In addition, simple nonlinear enhancement and a denosing stage that preserves edges using wavelet shrinkage are computed into this technique. In this paper. we propose a new method for the gain setting of nonlinear enhancement and show result and comparison.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.3
• /
• pp.173-180
• /
• 2004

In this paper we provide a new image restoration method based on the multiscale regularization in the redundant wavelet transform domain. The proposed method uses the redundant wavelet transform to decompose the single-scale image restoration problem to multiscale ones and applies scale dependent regularization to the decomposed restoration problems. The proposed method recovers sharp edges by applying rather less regularization to wavelet related restorations, while suppressing the resulting noise magnification by the wavelet shrinkage algorithm. The improved performance of the proposed method over more traditional Wiener filtering is shown through numerical experiments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.6C
• /
• pp.589-602
• /
• 2006

잡음은 음성과 관련된 시스템의 성능을 제한하는 주된 원인이기 때문에 음성향상과 관련된 연구는 꾸준히 계속되어왔다. 전통적인 음성향상 방법은 무성음과 잡음을 구분하지 알기 때문에 잡음제거 과정에서 무성음이 함께 제거되는 단점이 있으며, 웨이블릿 기반의 전통적인 잡음제거 방법은 각 대역마다 동일한 문턱값을 사용하기 때문에 시변 환경에서 성능이 떨어지는 단점이 있다. 이 단점들을 개선하기위해 다중대역 파워 차감법과 Perceptual 웨이블릿 패킷 분해를 이용한 웨이블릿 기반의 개선된 음성향상 방법을 제안한다. 전처리 과정으로 다중대역 파워 차감법을 사용하여 광대역 잡음을 제거하고 뮤지컬 잡음의 발생을 줄이며, psycho-acoustic 모델 기반 Perceptual 웨이블릿 패킷으로 신호를 분해한 후 각 웨이블릿 노드의 엔트로피 비율과 음성검출을 이용하여 무성음/유성음/잡음을 구분한다. 구분된 신호에 따라 각 웨이블릿 노드마다의 문턱값을 기준으로 웨이블릿 Shrinkage를 적용하여 잡음을 제거하고 무성음이나 파워가 작은 유성음이 제거되는 오류를 최소화한다. 또한 잡음 파워 추정 과정에 적응적으로 망각 계수를 선택하여 잡음 파워 추정 오류를 최소화한다.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.2
• /
• pp.467-476
• /
• 2004

Recently, an approximation of a wavelet series has been developed in the analyses of an unknown function. Most of articles have been studied on thresholding and shrinkage methods for its wavelet coefficients based on (non)parametric and Bayesian methods when the sample size is considered as a maximum resolution in wavelet series. In this paper, regardless of the sample size, we are focusing only on the choice of a maximum resolution in wavelet series. We propose a Bayesian approach to the choice of a maximum resolution based on the linear combination of the wavelet basis functions.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.5
• /
• pp.1-9
• /
• 2011

This paper presents a novel scratch detection and removal approach for old film images in wavelet-domain. Various scratch detection and removal algorithms have been proposed for past decades. However, accurate scratch detection and removal with a moderate amount of computing effort is still a problem because of trade off between the quality of the film and computational load. For overcoming this problem, we first decompose an input image using a 3-level wavelet transform, and then remove the scratch by shrinking wavelet coefficients using linear interpolation. Experimental results show that the proposed algorithm can efficiently detect and remove the scratch in damaged films, and also be incorporated into old film restoration systems.

• Journal of the Institute of Convergence Signal Processing
• /
• v.9 no.1
• /
• pp.10-17
• /
• 2008

The purpose of this thesis is to reduce of error for source location through acoustic emission(AE) signal, generated elastic wave from crack growth to leak for facility diagnosis. Especially, in order to overcome noise from original signal, this paper proposed enhancement of source location by using noise reduction based on wavelet transform. To evaluate actual performance in experiments, Pencil Lead Break is used crack signal source on the aluminum plate and drain valve of air compressor is used as substitute pressure vessel to generate leak signal. In signal processing, wavelet shrinkage and soft threshold are used to discriminate signal source and then source location techniques have been effectively used with group velocity using material property and time difference between sensor using cross correlation. Source location for crack and leak test have some difference, but the result show that improved 30% with a average length within 10.46mm in crack test and improved 2% compare with average filter in leak test when we applied wavelet transform.