Non-linear Diffusion Models for Despeckling of Images: Achievements and Future Challenges

Jain, Subit K.; Ray, Rajendra K.

doi:10.1080/02564602.2019.1565960

Cited by 18 publications

(15 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the literature [3,4], it was found that the edge and slope regions of an image can be identified by differential curvature, and both constructed two adaptive diffusion coefficients to discriminate the whole image. In the literature [5], a nonlinear anisotropic combined diffusion model was designed; this model is mainly by choosing different diffusion methods when facing different regions of the image, homogeneous diffusion for smooth regions, and mean curvature diffusion for edge regions, and the effectiveness of this model was experimentally demonstrated. The literature [6] combined the concept of fractional order to improve a model of anisotropic diffusion, which also led to the research on the application of fractional order in the field of image edge detection.…”

Section: Related Studiesmentioning

confidence: 99%

Fractional-Order Adaptive $P$ -Laplace Equation-Based Art Image Edge Detection

Chen

2021

Advances in Mathematical Physics

View full text Add to dashboard Cite

In recent years, with the rapid development of image processing research, the study of nonstandard images has gradually become a research hotspot, for example, fabric images, remote sensing images, and gear images. Some of the remote sensing images have a complex background and low illumination compared with standard images and are easy to be mixed with noise during acquisition; some of the fabric images have rich texture information, which adds difficulty to the related processing, and are also easy to be mixed with noise during acquisition. In this paper, we propose a fractional-order adaptive P -Laplace equation image edge detection algorithm for the problem of image edge detection in which the edge and texture information of the image is lost. The algorithm can apply for the order adaptively to filter the noise according to the noise distribution of the image, and the adaptive diffusion factor is determined by both the fractional-order curvature and fractional-order gradient of the iso-illumination line and combined with the iterative approach to realize the fine-tuning of the noisy image. The experimental results demonstrate that the algorithm can remove the noise while preserving the texture and details of the image. A fractional-order partial differential equation image edge detection model with a fractional-order fidelity term is proposed for Gaussian noise. The model incorporates a fractional-order fidelity term because this fidelity term smoothes out the rougher parts of the image while preserving the texture in the original image in greater detail and eliminating the step effect produced by other models such as the Perona-Malik (PM) and Rudin-Osher-Fatemi (ROF) models. By comparing with other algorithms, the image edge detection effect is measured with the help of evaluation metrics such as peak signal-to-noise ratio and structural similarity, and the optimal value is selected iteratively so that the image with the best edge detection result is retained. A convolutional mask image edge detection model based on adaptive fractional-order calculus is proposed for the scattered noise in medical images. The adaption is mainly reflected in the model algorithm by constructing an exponential parameter relation that is closely related to the image, which can dynamically adjust the parameter values, thus making the model algorithm more practical. The model achieves the scattering noise removal in four steps.

show abstract

Section: Related Studiesmentioning

confidence: 99%

Fractional-Order Adaptive $P$ -Laplace Equation-Based Art Image Edge Detection

Chen

2021

Advances in Mathematical Physics

View full text Add to dashboard Cite

show abstract

“…3, where the speckle noise cancellation (SNC) factors for each of the i-th frame g i (m, n) multiplies with h ei (m, n), to obtain an estimate of the true US image ri (m, n) for that frame. In the absense of additive noise in (13) and estimation error in the estimated speckle pattern Ûi , the estimated ultrasound image of the i-th frame, Ri , can be obtained in matrix form as…”

Section: True Imagementioning

confidence: 99%

“…Again, the nonlinear approaches based on the diffusion equation [12] not only preserves edges but also enhances edges by inhibiting diffusion across edges and allowing diffusion on either side of the edge. However, selection of the parameter-values is a major issue in this method, as a value of parameter that is smaller than the optimum one leads to unsatisfactory noise suppression whereas a higher parameter-value results in poor structure preservation [13].…”

Section: Introductionmentioning

confidence: 99%

Multiframe-based Adaptive Despeckling Algorithm for Ultrasound B-mode Imaging with Superior Edge and Texture

Dey,

Hasan

2019

Preprint

View full text Add to dashboard Cite

Removing speckle noise from medical ultrasound images while preserving image features without introducing artifact and distortion is a major challenge in ultrasound image restoration. In this paper, we propose a multiframe-based adaptive despeckling (MADS) algorithm to reconstruct a highresolution B-mode image from raw radio-frequency (RF) data that is based on a multiple input single output (MISO) model. As a prior step to despeckling, the speckle pattern in each frame is estimated using a novel multiframe-based adaptive approach for ultrasonic speckle noise estimation (MSNE) based on a single input multiple output (SIMO) modeling of consecutive deconvolved ultrasound image frames. The elegance of the proposed despeckling algorithm is that it addresses the despeckling problem by completely following the signal generation model unlike conventional ad-hoc smoothening or filtering based approaches, and therefore, it is likely to maximally preserve the image features. As deconvolution is a necessary pre-processing step to despeckling, we describe here a 2-D extension of the SIMO model-based 1-D deconvolution method. Finally, a complete framework for the generation of high-resolution ultrasound B-mode image has been also established in this paper. The results show 8.55 − 15.91 dB, 8.24 − 14.94 dB improvement in terms of SNR and PSNR, respectively, for simulation data and 2.22 − 3.17, 13.24 − 32.85 improvement in terms of NIQE and BRISQUE, respectively, for in-vivo data compared to the traditional despeckling algorithms. Visual comparison shows superior texture, resolution, details of B-mode images offered by our method compared to those by a commercial scanner, and hence, it may significantly improve the diagnostic quality of ultrasound images.

show abstract

“…This attraction is characterized by several modifications of the Perona-Malik model that are published in different journals (Guo et al, 2012;Kessy et al, 2017a;Kessy et al, 2017b;Maiseli et al, 2018). These works are establishing stable and accurate models that deal with different noise variants and staircase artifacts caused by the ill-posed aspect associated with the partial differentiation applied in the Perona-Malik kernel (Liu et al, 2013;Jain and Ray, 2019;Yao et al, 2019). In general, the Perona-Malik model is made of a diffusion kernel functional that approximates the pixel value and the regularization term, which has been added to control the illposed aspect of the model and prevents staircase artifacts in the despeckled image.…”

Section: Introductionmentioning

confidence: 99%

Comparative Analysis of Multiplicative and Additive Noise Based Automated Regularizations in Non-Linear Diffusion Image Reconstruction

Christophe

Anthony

Kapyela

et al. 2020

TJET

View full text Add to dashboard Cite

Multiplicative and additive noises are often introduced in image signals during the image acquisition process and result into degradation of image features. The work done by Perona and Malik in 1990 and its modified versions revolutionized the way through which noises or speckles are removed. The Perona-Malik model requires tuning of the regularization parameter to control and prevent staircase artifacts in restored images. The current manual tuning is a challenging and time consuming practice when a long queue of images is registered for processing. Attempt to automate the regularization parameter appeared in Perona-Malik model with self-adjusting shape-defining constant. Although both multiplicative and additive noise based automated regularizations were presented, the paper stayed silent on matters concerning the automation method that fits with speckle reduction. This paper therefore, presents a comparative analysis of additive and multiplicative noise based automated regularizations. Simulation results and paired samples T-tests reveal that the multiplicative noise based automation outperforms the additive noise based automation for small speckle variances. However, the two automation methods do not significantly differ when large speckle variances are assumed.

show abstract

Non-linear Diffusion Models for Despeckling of Images: Achievements and Future Challenges

Cited by 18 publications

References 65 publications

Fractional-Order Adaptive $P$ -Laplace Equation-Based Art Image Edge Detection

Fractional-Order Adaptive $P$ -Laplace Equation-Based Art Image Edge Detection

Multiframe-based Adaptive Despeckling Algorithm for Ultrasound B-mode Imaging with Superior Edge and Texture

Comparative Analysis of Multiplicative and Additive Noise Based Automated Regularizations in Non-Linear Diffusion Image Reconstruction

Contact Info

Product

Resources

About

Non-linear Diffusion Models for Despeckling of Images: Achievements and Future Challenges

Cited by 18 publications

References 65 publications

Fractional-Order Adaptive P -Laplace Equation-Based Art Image Edge Detection

Fractional-Order Adaptive P -Laplace Equation-Based Art Image Edge Detection

Multiframe-based Adaptive Despeckling Algorithm for Ultrasound B-mode Imaging with Superior Edge and Texture

Comparative Analysis of Multiplicative and Additive Noise Based Automated Regularizations in Non-Linear Diffusion Image Reconstruction

Contact Info

Product

Resources

About

Fractional-Order Adaptive $P$ -Laplace Equation-Based Art Image Edge Detection

Fractional-Order Adaptive $P$ -Laplace Equation-Based Art Image Edge Detection