Multiscale Nonlinear Diffusion and Shock Filter for Ultrasound Image Enhancement

Anisotropic diffusion based on directional consistency

“…Chen proposed contextual adaptive smoothing algorithm [7], both local inhomogeneity and spatial gradient were used in diffusion, different scale was used in different contextual. Another multi-scale diffusion algorithm is Zhang's Laplacian pyramid-based nonlinear diffusion and shock filter (LPNDSF) [8]. Wang's local variance-controlled forward and-backward (LVCFAB) diffusion combines Chen's adaptive smoothing with F AB [9].…”

Multiscale adaptive regularisation Savitzky–Golay method for speckle noise reduction in ultrasound images

“…Thus, multiscale analysis can be applied to separate noise from meaningful details. Although several image decomposition techniques [10][11][12][13][14][15][16][17][18][19][20][21][22][23] have been employed for speckle noise reduction, most techniques are still based on the classic Laplacian pyramid [24] and wavelet transform [25]. From our evaluation of the denoising techniques based on wavelet transform [11][12][13][14][15][16][17], nonlinear multiscale wavelet diffusion (NMWD) method [11] provided the best result in terms of speckle noise suppression and edge enhancement; however, it removes some details.…”

“…From our evaluation of the denoising techniques based on wavelet transform [11][12][13][14][15][16][17], nonlinear multiscale wavelet diffusion (NMWD) method [11] provided the best result in terms of speckle noise suppression and edge enhancement; however, it removes some details. In the same setting, we found that feature-enhanced speckle reduction (FESR) method [23] is the best denoising technique based on Laplacian pyramid among [19][20][21][22][23]. In FESR method, the position of the edge can be shifted which may lead to incorrect measurement.…”

“…In this paper, we propose the denoising algorithm that provides the good balance between signal smoothing (denoising) and structural preservation. The majority of denoising techniques for medical ultrasound images decompose an image into four layers [11,15,19,20,23]; however, the researchers of [27,28] argue that three layers are sufficient for effective denoising. We adapt the decomposition of [24] and separate the image into three layers: noise, texture and object layers.…”

Enhancement of the ultrasound images by modified anisotropic diffusion method

“…Apart from anisotropic diffusion methods, several multiscale approaches [8,14,15,26,28,30,32,42,45] were also proposed to reduce speckle in ultrasound images. Most of the recent studies [22,23,27,[43][44][45] on speckle reduction techniques are based on fusion of anisotropic diffusion and multiscale techniques. These techniques are more suited for segmentation purposes rather than feature extraction and visual diagnosis from the real ultrasound images due to the complexity of speckle statistics involved in feature extraction and visual diagnosis.…”