A Multiscale Neural Network Method for Image Restoration

Castro, A.P.A. de; Drummond, Isabela; Silva, J.D.S. da

doi:10.5540/tema.2008.09.01.0041

Cited by 5 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previously, non-medical images, such as images of human faces, vegetables, and birds have been denoised with neural network [ 12 , 21 , 22 ], and multilayer perceptron or convolutional neural network have achieved good performance for image denoising of non-medical images. For medical image denoising, there are several studies to use conventional neural network [ 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 ]. On the other hand, literature survey of Litjens et al shows that, in deep learning with convolutional neural network or CAE, number of applications of image enhancement like image denoising is limited [32] .…”

Section: Discussionmentioning

confidence: 99%

Convolutional auto-encoder for image denoising of ultra-low-dose CT

Nishio

Nagashima

Hirabayashi

et al. 2017

Heliyon

View full text Add to dashboard Cite

ObjectivesThe purpose of this study was to validate a patch-based image denoising method for ultra-low-dose CT images. Neural network with convolutional auto-encoder and pairs of standard-dose CT and ultra-low-dose CT image patches were used for image denoising. The performance of the proposed method was measured by using a chest phantom.Materials and methodsStandard-dose and ultra-low-dose CT images of the chest phantom were acquired. The tube currents for standard-dose and ultra-low-dose CT were 300 and 10 mA, respectively. Ultra-low-dose CT images were denoised with our proposed method using neural network, large-scale nonlocal mean, and block-matching and 3D filtering. Five radiologists and three technologists assessed the denoised ultra-low-dose CT images visually and recorded their subjective impressions of streak artifacts, noise other than streak artifacts, visualization of pulmonary vessels, and overall image quality.ResultsFor the streak artifacts, noise other than streak artifacts, and visualization of pulmonary vessels, the results of our proposed method were statistically better than those of block-matching and 3D filtering (p-values < 0.05). On the other hand, the difference in the overall image quality between our proposed method and block-matching and 3D filtering was not statistically significant (p-value = 0.07272). The p-values obtained between our proposed method and large-scale nonlocal mean were all less than 0.05.ConclusionNeural network with convolutional auto-encoder could be trained using pairs of standard-dose and ultra-low-dose CT image patches. According to the visual assessment by radiologists and technologists, the performance of our proposed method was superior to that of large-scale nonlocal mean and block-matching and 3D filtering.

show abstract

Section: Discussionmentioning

confidence: 99%

Convolutional auto-encoder for image denoising of ultra-low-dose CT

Nishio

Nagashima

Hirabayashi

et al. 2017

Heliyon

View full text Add to dashboard Cite

show abstract

“…Debakla, Djemal, & Benyettou [28] proposed a multilayer neural network to reduce the total variation due to noise exposure the image. The ideal mainly based on weights in neural network to reduce appropriate functional and get optimal solution, the proposed model exhibited high performance restoring geometric properties such as corners and edges, the result outperformed the Tichonov regularization, ROF proposed by Rudin, Osher, and Fatemi [29] and Multiscale Neural Network suggested by Castro, Drummond, and Silva [30]. The last mentioned model, used the ANN and local spatial information, composed two phase modified Kohonen neural network to clusterize the training data set, then the multilayer neural network used to recover the inverse reconstruction model.…”

Section: B Ann In Image Processingmentioning

confidence: 95%

The Impact of Neural Network Techniques in the Optimization of the Image Processing

Anwer¹

2020

Journal of Southwest Jiaotong University

View full text Add to dashboard Cite

Nowadays, image processing is widely utilized in many applications and for various purposes. Scholars proposed and suggested various techniques of image processing. The neural network is one of the main processing techniques, which is a state-of-art method. This paper aims to investigate neural network techniques in the field of image processing. Moreover, medical imaging, as well as increasing trends of utilizing digital medical imaging, has gained huge attention in the health sectors. In this regard, this paper focuses on the effect of neural networks in optimizing medical image processing. In this context, the early diagnosis and detection of the eye have an important role in the avoidance of visual impairment, because of the fact that around 45 million people have visual impairments all over the world, according to the World Health Organization. For this reason, the current paper introduces a new method based on image processing for vascular segmentation based on a morphological active contour. Then, segmentation carried out based on morphological operations, fuzzy c-means, and watershed transform. The output of such segmentation methods was given to conventional neural network. The optimized feature values are then extracted. The threshold value is set to compare these optimized feature values. From this, the best segmentation methods will be obtained.

show abstract

“…Castro et al [17] proposed to use a multiscale neural network approach for restoring degraded images.…”

Section: Related Workmentioning

confidence: 99%

“…All terms of the equations (16) and (17) are calculated from the value of i w u ∂ ∂ and i c u ∂ ∂ so it is necessary to provide firstly these derivatives. Under our assumptions and considerations in (9) and as N(x, y) is considered in (10), under these conditions, the derivatives can be given as follows:…”

Section: Training Of Rbfnnfmentioning

confidence: 99%

A Novel Approach for Medical Images Noise Reduction Based RBF Neural Network Filter

Salem¹,

Djemal

Benyettou³

2015

JCP

View full text Add to dashboard Cite

This paper is dedicated to the presentation of a Radial basis function neural network (RBFNN) based denoising method for medical images. In the proposed approach, a RBFNN filter is designed where the output of the network is a single denoised pixel and the inputs are its neighborhood in the degraded image. The back-propagation algorithm is used to train the RBFNN filter by minimizing an appropriate error function obtained from the total variation model. The parameters to be adjusted are the weights and the neurons centers of the RBFNN. The considered filter was used to reduce noise from X-ray, MRI and Mammographic medical images giving good results of noise removal when compared to other approaches and using different noise standard deviations.

show abstract

A Multiscale Neural Network Method for Image Restoration

Cited by 5 publications

References 9 publications

Convolutional auto-encoder for image denoising of ultra-low-dose CT

Convolutional auto-encoder for image denoising of ultra-low-dose CT

The Impact of Neural Network Techniques in the Optimization of the Image Processing

A Novel Approach for Medical Images Noise Reduction Based RBF Neural Network Filter

Contact Info

Product

Resources

About