Bayesian denoising in digital radiography: A comparison in the dental field

Frosio, Iuri; Olivieri, Chiara; Lucchese, M. G.; Borghese, N. Alberto; Boccacci, Patrizia

doi:10.1016/j.compmedimag.2012.10.003

Cited by 6 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We typically choose a cost e = E {õ j }, {o j }; θ 0 , · · · , θ P−1 related to the quality of the processed images (for instance, [2] uses the average PSNR of {õ j }). Computing the derivatives of e with respect to the processing parameters may be difficult or even impossible in certain cases: for instance, when the parameters are discrete, as in the case of the patch size for NLM; or when the processing algorithm is iterative, as for TV denoising [6,16]; or when the cost function is not differentiable, as in the case of FSIM [17]. Therefore, we resort to a derivative-free optimization algorithm, the Nelder-Mead simplex method [18].…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Learning Adaptive Parameter Tuning for Image Processing

Dong¹,

Frosio²,

Kautz³

2018

View full text Add to dashboard Cite

The non-stationary nature of image characteristics calls for adaptive processing, based on the local image content. We propose a simple and flexible method to learn local tuning of parameters in adaptive image processing: we extract simple local features from an image and learn the relation between these features and the optimal filtering parameters. Learning is performed by optimizing a user defined cost function (any image quality metric) on a training set. We apply our method to three classical problems (denoising, demosaicing and deblurring) and we show the effectiveness of the learned parameter modulation strategies. We also show that these strategies are consistent with theoretical results from the literature.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The last application we consider is image deblurring through TV regularization. This problem is of particular importance in the medical and astronomical fields, where imaging apparatuses with a known point spread function measure single channel images with a very limited number of photons [6,16]. The problem can be stated as follow.…”

Section: Image Deblurringmentioning

confidence: 99%

Learning Adaptive Parameter Tuning for Image Processing

Dong¹,

Frosio²,

Kautz³

2018

View full text Add to dashboard Cite

show abstract

“…In addition, a subjective visual assessment of the images was performed to appreciate the practical performance of the denoising methods from the clinicians' point of view. To our knowledge, very few studies 26,27 have investigated the denoising of dental radiographs, and in fact, the current study is the first to propose and evaluate the application of the BM3D and total variation methods for denoising dental micro-CT images.…”

Section: Discussionmentioning

confidence: 99%

A comparative study of new and current methods for dental micro-CT image denoising

Shahmoradi¹,

Lashgari²,

Rabbani³

et al. 2016

Dentomaxillofacial Radiology

View full text Add to dashboard Cite

Objectives:The aim of the current study was to evaluate the application of two advanced noise-reduction algorithms for dental micro-CT images and to implement a comparative analysis of the performance of new and current denoising algorithms. Methods: Denoising was performed using gaussian and median filters as the current filtering approaches and the block-matching and three-dimensional (BM3D) method and total variation method as the proposed new filtering techniques. The performance of the denoising methods was evaluated quantitatively using contrast-to-noise ratio (CNR), edge preserving index (EPI) and blurring indexes, as well as qualitatively using the double-stimulus continuous quality scale procedure. Results: The BM3D method had the best performance with regard to preservation of fine textural features (CNR Edge ), non-blurring of the whole image (blurring index), the clinical visual score in images with very fine features and the overall visual score for all types of images. On the other hand, the total variation method provided the best results with regard to smoothing of images in texture-free areas (CNR Tex-free ) and in preserving the edges and borders of image features (EPI). Conclusions: The BM3D method is the most reliable technique for denoising dental micro-CT images with very fine textural details, such as shallow enamel lesions, in which the preservation of the texture and fine features is of the greatest importance. On the other hand, the total variation method is the technique of choice for denoising images without very fine textural details in which the clinician or researcher is interested mainly in anatomical features and structural measurements. Dentomaxillofacial Radiology (2016Radiology ( ) 45, 20150302. doi: 10.1259 Cite this article as: Shahmoradi M, Lashgari M, Rabbani H, Qin J, Swain M. A comparative study of new and current methods for dental micro-CT image denoising. Dentomaxillofac Radiol 2016; 45: 20150302.

show abstract

“…These categories are denoising, contrast enhancement, and sharpening algorithms. The denoising algorithms include a 2-D Butterworth low-pass filter (frequency domain) [9], Bayesian Least Squares -Gaussian Scaled Mixture (BLS-GSM) algorithm, and Total Variation (TV) filter [10]. Contrast Limited Adaptive Histogram Equalization (CLAHE) [11,12] is an algorithm in literature meant for improving the contrast in dental radiographs.…”

Section: Review Of Literaturementioning

confidence: 99%

“…As mentioned, each value in the 2D vector corresponding to the difference, d is weighted by the corresponding value in the normalized gradient vector as, The operator, Γ in (9) indicates element-wise multiplication. The sharpened radiograph, (10) (11) In (10), α is an arbitrary parameter that determines the strength of sharpening. This parameter is called ad the scale.…”

Section: Gradient-adaptive Nonlinear Sharpening (Gns)mentioning

confidence: 99%

Gradient-adaptive Nonlinear Sharpening for Dental Radiographs

Joy

Titus²

2023

Int. j. electr. comput. eng. syst. (Online)

View full text Add to dashboard Cite

Unsharp Masking is a popular image processing technique used for improving the sharpness of structures on dental radiographs. However, it produces overshoot artefact and intolerably amplifies noise. On radiographs, the overshoot artefact often resembles the indications of prosthesis misfit, pathosis, and pathological features associated with restorations. A noise- robust alternative to the Unsharp Masking algorithm, termed Gradient-adaptive Nonlinear Sharpening (GNS) which is free from overshoot and discontinuity artefacts, is proposed in this paper. In GNS, the product of the arbitrary scalar termed as ‘scale’ and the difference between the output of the Adaptive Edge Smoothing Filter (AESF) and the input image, weighted by the normalized gradient magnitude is added to the input image. AESF is a locally-adaptive 2D Gaussian smoothing kernel whose variance is directly proportional to the local value of the gradient magnitude. The dataset employed in this paper is downloaded from the Mendeley data repository having annotated panoramic dental radiographs of 116 patients. On 116 dental radiographs, the values of Saturation Evaluation Index (SEI), Sharpness of Ridges (SOR), Edge Model Based Contrast Metric (EMBCM), and Visual Information Fidelity (VIF) exhibited by the Unsharp Masking are 0.0048 ± 0.0021, 4.4 × 1013 ± 3.8 × 1013, 0.2634 ± 0.2732 and 0.9898 ± 0.0122. The values of these quality metrics corresponding to the GNS are 0.0042 ± 0.0017, 2.2 × 1013 ± 1.8 × 1013, 0.5224 ± 0.1825, and 1.0094 ± 0.0094. GNS exhibited lower values of SEI and SOR and higher values of EMBCM and VIF, compared to the Unsharp Masking. Lower values of SEI and SOR, respectively indicate that GNS is free from overshoot artefact and saturation and the quality of edges in the output images of GNS is less affected by noise. Higher values of EMBCM and VIF, respectively confirm that GNS is free from haloes as it produces thin and sharp edges and the sharpened images are of good information fidelity.

show abstract

Bayesian denoising in digital radiography: A comparison in the dental field

Cited by 6 publications

References 28 publications

Learning Adaptive Parameter Tuning for Image Processing

Learning Adaptive Parameter Tuning for Image Processing

A comparative study of new and current methods for dental micro-CT image denoising

Gradient-adaptive Nonlinear Sharpening for Dental Radiographs

Contact Info

Product

Resources

About