Analysis the Statistical Parameters of the Wavelet Coefficients for Image Denoising.pdf
Được đăng lên bởi
Nguyễn Khắc Hấn
Số trang: 7 trang

Lượt xem: 406 lần

Lượt tải: 0 lần
VNU Journal of Natural Sciences and Technology, Vol. 29, No. 3 (2013) 17 Analysis the Statistical Parameters of the Wavelet Coefficients for Image Denoising Nguyễn Vĩnh An* PetroVietnam University, 173 Trung Kính, Cầu Giấy, Hanoi, Vietnam Received 14 September 2012 Revised 28 September 2012; accepted 28 June 2013 Abstract: Image denoising is aimed at the removal of noise which may corrupt an image during its acquisition or transmission. Denoising of the corrupted image by Gaussian noise using wavelet transform is very effective way because of its ability to capture the energy of a signal in few larger values. This paper proposes a threshold selection method for image denoising based on the statistical parameters which depended on subband data. The threshold value is computed based on the number of coefficients in each scale j of wavelet decomposition and the noise variance in various subband. Experimental results in PSNR on several test images are compared for different denoise techniques. 1. Introduction∗ (intensity spikes). Speckle noise is multiplicative noise which occurs in almost all coherent systems. Image denoising is still a challenging problem for researchers as which causes blurring and introduces artifacts. Denoising method tends to be problem specific and depends upon the type of image and noise model. Image denoising is a common procedure in digital image processing aiming at the removal of noise which may corrupt an image during its acquisition or transmission while sustaining its quality. Noise is unwanted signal that interferes with the original signal and degrades the quality of the digital image. Different types of images inherit different types of noise and different noise models are used for different noise types. Denoising based on transform domain filtering and wavelet can be subdivided into data adaptive and nonadaptive filters [2]. Image denoising based on spatial domain filtering is classified into linear filters and nonlinear filters [3, 4]. In [5, 6], the paper proposes an adaptive, data driven threshold for image denoising via wavelet soft thresholding. Noise is present in image either in additive or multiplicative form [1]. Various types of noise have their own characteristics and are inherent in images in different ways. Gaussian noise is evenly distributed over the signal. Salt and pepper noise is an impulse type of noise _______ ∗ A proposal of vector/matrix extension of denoising algorithm developed for ...
VNU Journal of Natural Sciences and Technology, Vol. 29, No. 3 (2013) 17
1
Analysis the Statistical Parameters of the Wavelet Coefficients
for Image Denoising
Nguyễn Vĩnh An*
PetroVietnam University, 173 Trung Kính, Cầu Giấy, Hanoi, Vietnam
Received 14 September 2012
Revised 28 September 2012; accepted 28 June 2013
Abstract: Image denoising is aimed at the removal of noise which may corrupt an image during
its acquisition or transmission. Denoising of the corrupted image by Gaussian noise using wavelet
transform is very effective way because of its ability to capture the energy of a signal in few larger
values. This paper proposes a threshold selection method for image denoising based on the
statistical parameters which depended on subband data. The threshold value is computed based on
the number of coefficients in each scale j of wavelet decomposition and the noise variance in
various subband. Experimental results in PSNR on several test images are compared for different
denoise techniques.
1. Introduction
∗
∗∗
∗
Image denoising is a common procedure in
digital image processing aiming at the removal
of noise which may corrupt an image during its
acquisition or transmission while sustaining its
quality. Noise is unwanted signal that interferes
with the original signal and degrades the quality
of the digital image. Different types of images
inherit different types of noise and different
noise models are used for different noise types.
Noise is present in image either in additive
or multiplicative form [1]. Various types of
noise have their own characteristics and are
inherent in images in different ways. Gaussian
noise is evenly distributed over the signal. Salt
and pepper noise is an impulse type of noise
_______
∗
Tel: 84913508067.
Email: annv@pvu.edu.vn
(intensity spikes). Speckle noise is
multiplicative noise which occurs in almost all
coherent systems. Image denoising is still a
challenging problem for researchers as which
causes blurring and introduces artifacts. Denoising
method tends to be problem specific and depends
upon the type of image and noise model.
Denoising based on transform domain
filtering and wavelet can be subdivided into
data adaptive and nonadaptive filters [2].
Image denoising based on spatial domain
filtering is classified into linear filters and non
linear filters [3, 4]. In [5, 6], the paper proposes
an adaptive, data driven threshold for image
denoising via wavelet soft thresholding.
A proposal of vector/matrix extension of
denoising algorithm developed for grayscale
images, in order to efficiently process
Để xem tài liệu đầy đủ. Xin vui lòng
Đăng nhập
Nếu xem trực tuyến bị lỗi, bạn có thể tải về máy để xem.
Analysis the Statistical Parameters of the Wavelet Coefficients for Image Denoising.pdf

Người đăng:
Nguyễn Khắc Hấn
5
Tài liệu rất hay!
Được đăng lên bởi
Mai Hương

1 giờ trước
Đúng là cái mình đang tìm. Rất hay và bổ ích. Cảm ơn bạn!
7
Vietnamese
Analysis the Statistical Parameters of the Wavelet Coefficients for Image Denoising.pdf
9
10
16