scholarly journals Median Filter Aided CNN Based Image Denoising: An Ensemble Aprroach

Algorithms ◽  
2021 ◽  
Vol 14 (4) ◽  
pp. 109
Author(s):  
Subhrajit Dey ◽  
Rajdeep Bhattacharya ◽  
Friedhelm Schwenker ◽  
Ram Sarkar
Keyword(s):  
Neural Networks ◽  
Image Denoising ◽  
Gaussian Noise ◽  
Additive White Gaussian Noise ◽  
Median Filter ◽  
Research Problem ◽  
The Third ◽  
Challenging Research ◽  
Deep Cnn ◽  
Dilation Rate

Image denoising is a challenging research problem that aims to recover noise-free images from those that are contaminated with noise. In this paper, we focus on the denoising of images that are contaminated with additive white Gaussian noise. For this purpose, we propose an ensemble learning model that uses the output of three image denoising models, namely ADNet, DnCNN, and IRCNN, in the ratio of 2:3:6, respectively. The first model (ADNet) consists of Convolutional Neural Networks with attention along with median filter layers after every convolutional layer and a dilation rate of 8. In the case of the second model, it is a feed forward denoising CNN or DnCNN with median filter layers after half of the convolutional layers. For the third model, which is Deep CNN Denoiser Prior or IRCNN, the model contains dilated convolutional layers and median filter layers up to the dilated convolutional layers with a dilation rate of 6. By quantitative analysis, we note that our model performs significantly well when tested on the BSD500 and Set12 datasets.

scholarly journals SAR Image Despeckling by Deep Neural Networks: from a Pre-Trained Model to an End-to-End Training Strategy

2020 ◽  
Vol 12 (16) ◽  
pp. 2636
Author(s):  
Emanuele Dalsasso ◽  
Xiangli Yang ◽  
Loïc Denis ◽  
Florence Tupin ◽  
Wen Yang
Keyword(s):  
Neural Networks ◽  
Gaussian Noise ◽  
State Of The Art ◽  
Additive White Gaussian Noise ◽  
Hybrid Approach ◽  
White Gaussian Noise ◽  
Training Data ◽  
Speckle Reduction ◽  
Sar Image ◽  
Sar Images

Speckle reduction is a longstanding topic in synthetic aperture radar (SAR) images. Many different schemes have been proposed for the restoration of intensity SAR images. Among the different possible approaches, methods based on convolutional neural networks (CNNs) have recently shown to reach state-of-the-art performance for SAR image restoration. CNN training requires good training data: many pairs of speckle-free/speckle-corrupted images. This is an issue in SAR applications, given the inherent scarcity of speckle-free images. To handle this problem, this paper analyzes different strategies one can adopt, depending on the speckle removal task one wishes to perform and the availability of multitemporal stacks of SAR data. The first strategy applies a CNN model, trained to remove additive white Gaussian noise from natural images, to a recently proposed SAR speckle removal framework: MuLoG (MUlti-channel LOgarithm with Gaussian denoising). No training on SAR images is performed, the network is readily applied to speckle reduction tasks. The second strategy considers a novel approach to construct a reliable dataset of speckle-free SAR images necessary to train a CNN model. Finally, a hybrid approach is also analyzed: the CNN used to remove additive white Gaussian noise is trained on speckle-free SAR images. The proposed methods are compared to other state-of-the-art speckle removal filters, to evaluate the quality of denoising and to discuss the pros and cons of the different strategies. Along with the paper, we make available the weights of the trained network to allow its usage by other researchers.

Image denoising via Bayesian estimation of local variance with Maxwell density prior

2015 ◽  
Vol 06 (02) ◽  
pp. 1550002
Author(s):  
Pichid Kittisuwan
Keyword(s):  
Image Denoising ◽  
Gaussian Noise ◽  
Prior Distribution ◽  
Additive White Gaussian Noise ◽  
Maximum A Posteriori ◽  
Map Estimation ◽  
A Posteriori ◽  
Bayesian Techniques ◽  
Local Variance ◽  
Selection Of

The need for efficient image denoising methods has grown with the massive production of digital images and movies of all kinds. The distortion of images by additive white Gaussian noise (AWGN) is common during its processing and transmission. This paper is concerned with dual-tree complex wavelet-based image denoising using Bayesian techniques. Indeed, one of the cruxes of the Bayesian image denoising algorithms is to estimate the local variance of the image. Here, we employ maximum a posteriori (MAP) estimation to calculate local observed variance with Maxwell density prior for local observed variance and Gaussian distribution for noisy wavelet coefficients. Evidently, our selection of prior distribution is motivated by analytical and computational tractability. The experimental results show that the proposed method yields good denoising results.

scholarly journals Improved Image Denoising Methodology using Deep CNN Bilateral Filter Compared to Additional Methods

2021 ◽  
Vol 10 (5) ◽  
pp. 191-196
Author(s):  
Pallavi Bora ◽  
Kapil Chaudhary
Keyword(s):  
Standard Deviation ◽  
Comparative Study ◽  
Image Denoising ◽  
Gaussian Noise ◽  
Bilateral Filter ◽  
Filter Method ◽  
Novel Approach ◽  
Deep Cnn ◽  
Test Result

Image Denoising techniques are widely used to remove the noise from the images. Due to the ease of the bilateral filter, it is used very often to remove the noise from the images. In this paper, a novel approach has been proposed to enhanced bilateral filter in conjunction with CNN as a booster to eliminate Gaussian noise from Grey images. Studies reveal that standard CNN using a bilateral filter is the best technique to eliminate Gaussian noise from images along with high PSNR values. This paper also performs a comparative study of the various existing techniques for image denoising with the CNN technique and the applied Bilateral filter Method as a de facto to improve the results in terms of enhanced PSNR values. ECND Net (Enhanced CNN) applied to noisy images with standard deviation σ = 15 gives PSNR values up to 32.81 In comparison to this when both bilateral filter and deep CNN applied, in conjunction produces improved PSNR values up to 34.73 along with the equivalent standard deviation. The results in this work reveal better performance in terms of PSNR as compared to other methods. The test result proves that the bilateral filter Method along with CNN can improve the quality of restored images significantly better.

A Non-Local Image Denoising Technique Using Adaptive Filter Parameter

2014 ◽  
Vol 556-562 ◽  
pp. 4839-4842
Author(s):  
Song Yuan Tang
Keyword(s):  
Image Denoising ◽  
Gaussian Noise ◽  
Additive White Gaussian Noise ◽  
Wiener Filter ◽  
White Gaussian Noise ◽  
Filter Parameter ◽  
Iterative Computation ◽  
Local Mean ◽  
Non Local

This paper proposes a method to obtain the optimal filter parameter of the non-local mean (NLM) algorithm. The parameter is assumed to be a function of the variance of the additive white Gaussian noise and is adaptive estimated. The initialization of the variance of the additive white Gaussian noise is estimated by Wiener filter. Then the NLM filter is used to adaptively estimate the noise variance. The image denoising is an iterative computation till the parameter convergence. Experiments show that the proposed method can improve the quality of the denoised images efficiently.

scholarly journals EDGE–PRESERVING DENOISING BASED ON DYNAMIC PROGRAMMING ON THE FULL SET OF ADJACENCY GRAPHS

2017 ◽  
Vol XLII-2/W4 ◽  
pp. 55-60
Author(s):  
P. C. Thang ◽  
A. V. Kopylov ◽  
S. D. Dvoenko
Keyword(s):  
Dynamic Programming ◽  
Image Denoising ◽  
Gaussian Noise ◽  
Medical Diagnosis ◽  
Additive White Gaussian Noise ◽  
Experimental Results ◽  
Crucial Importance ◽  
High Quality ◽  
Edge Preserving ◽  
Adjacency Graphs

The ability of a denoising procedure to preserve fine image structures when suppressing unwanted noise has crucial importance for an accurate and effective medical diagnosis. We introduce here a new procedure of edge-preserving denoising for medical images, that combines the flexibility in prior assumptions, and computational effectiveness of parametric multi-quadratic dynamic programming with the increased accuracy of a tree-like representation of a discrete lattice based on the full set of possible adjacency graphs of image elements. Proposed procedure can effectively remove an additive white Gaussian noise with high quality. We provide experimental results in image denoising as well as comparison with related methods.

scholarly journals Iterative based image and video denoising by fractional block matching and transform domain filtering

2021 ◽  
Author(s):  
Amir Mehdizadeh Hemat Abadi ◽  
Mohammad Reza Hosseiny Fatemi
Keyword(s):  
Image Denoising ◽  
Gaussian Noise ◽  
Additive White Gaussian Noise ◽  
Weighted Average ◽  
Block Matching ◽  
Objective Measures ◽  
Transform Domain ◽  
Video Denoising ◽  
Simulation Results ◽  
Fractional Motion Estimation

This paper presents an iterative algorithm for image and video denoising which is based on fractional block-matching and transform domain filtering. We propose fractional motion estimation technique to find the most accurate similar blocks for each block of an image which improves sparsity enabling effective image denoising. By taking the advantage of blocks similarity and wavelet transform domain filtering along with weighted average function (WAF) in an iterative based manner, we achieve a higher level of sparsity and a better exploiting of blocks similarity redundancies of noisy images that increase the chance of preserving details and edges in the restored image. Since our algorithm is iterative, we can tradeoff between image denoising degree and computational complexity. In addition, we develop a video denoising algorithm based on the proposed image denoising algorithm. The simulation results of images and videos contaminated by additive white Gaussian noise demonstrate that our algorithm substantially achieves better denoising performance compared with previously published algorithms in terms of subjective and objective measures.

scholarly journals Edge-Preserving Median Filter and Weighted Coding with Sparse Nonlocal Regularization for Low-Dose CT Image Denoising Algorithm

2021 ◽  
Vol 2021 ◽  
pp. 1-7
Author(s):  
Quan Yuan ◽  
Zhenyun Peng ◽  
Zhencheng Chen ◽  
Yanke Guo ◽  
Bin Yang ◽  
...  
Keyword(s):  
Image Denoising ◽  
Gaussian Noise ◽  
Impulse Noise ◽  
Low Dose ◽  
Median Filter ◽  
Similarity Index ◽  
Ct Image ◽  
Edge Preserving ◽  
Low Dose Ct ◽  
Nonlocal Regularization

The impulse noise in CT image was removed based on edge-preserving median filter algorithm. The sparse nonlocal regularization algorithm weighted coding was used to remove the impulse noise and Gaussian noise in the mixed noise, and the peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) were calculated to evaluate the quality of the denoised CT image. It was found that in nine different proportions of Gaussian noise and salt-and-pepper noise in Shepp-Logan image and CT image processing, the PSNR and SSIM values of the proposed denoising algorithm based on edge-preserving median filter (EP median filter) and weighted encoding with sparse nonlocal regularization (WESNR) were significantly higher than those of using EP median filter and WESNR alone. It was shown that the weighted coding algorithm based on edge-preserving median filtering and sparse nonlocal regularization had potential application value in low-dose CT image denoising.

scholarly journals Iterative based image and video denoising by fractional block matching and transform domain filtering

2021 ◽  
Author(s):  
Amir Mehdizadeh Hemat Abadi ◽  
Mohammad Reza Hosseiny Fatemi
Keyword(s):  
Image Denoising ◽  
Gaussian Noise ◽  
Additive White Gaussian Noise ◽  
Weighted Average ◽  
Block Matching ◽  
Objective Measures ◽  
Transform Domain ◽  
Video Denoising ◽  
Simulation Results ◽  
Fractional Motion Estimation

This paper presents an iterative algorithm for image and video denoising which is based on fractional block-matching and transform domain filtering. We propose fractional motion estimation technique to find the most accurate similar blocks for each block of an image which improves sparsity enabling effective image denoising. By taking the advantage of blocks similarity and wavelet transform domain filtering along with weighted average function (WAF) in an iterative based manner, we achieve a higher level of sparsity and a better exploiting of blocks similarity redundancies of noisy images that increase the chance of preserving details and edges in the restored image. Since our algorithm is iterative, we can tradeoff between image denoising degree and computational complexity. In addition, we develop a video denoising algorithm based on the proposed image denoising algorithm. The simulation results of images and videos contaminated by additive white Gaussian noise demonstrate that our algorithm substantially achieves better denoising performance compared with previously published algorithms in terms of subjective and objective measures.

Sign in / Sign up

Export Citation Format

Share Document