Speckle Noise In Images Research Articles

Despeckling for side-scan sonar images based on adaptive block-matching and 3D filtering

Side-scan sonar (SSS) is an electronic device that utilizes the propagation characteristics of sound waves under water to complete underwater detection. Because the SSS produces images and maps according to the intensity of acoustic echo, speckle noise will be inevitably involved. A speckle denoising method based on block-matching and 3D filtering (BM3D) is proposed to filter the multiplicative speckle noise in SSS images. First, the SSS image is transformed by power and logarithm. The wavelet transform is used to estimate the general noisy level of the polluted image. Second, the parameters of the BM3D algorithm are updated according to the noise estimation results of each local patch. At last, after comparing the general noise estimation and the local noise estimation, the proposed algorithm chooses the best estimation to filter every patch separately to solve the problem that the noise is not evenly distributed. The experimental results show that the improved BM3D algorithm can effectively reduce the speckle noise in SSS images and obtain good visual effects. The Equivalent Number of Looks of the proposed algorithm is at least 6.83% higher, the Speckle Suppression Index is lower than traditional algorithm, and the Speckle Suppression and Mean Preservation Index is reduced by at least 3.30%. This method is mainly used for sonar image noise reduction, and has certain practical values for ultrasonic, radar or OCT images polluted by speckle noise.

Multiplicative-additive despeckling in SAR images

Visual and automatic analyses using synthetic aperture radar (SAR) images are challenging because of inherently formed speckle noise. Thus, reducing speckle noise in SAR images is an important research area for SAR image analysis. During speckle noise reduction, homogeneous regions should be smoothed while details such as edges and point scatterers need to be preserved. General speckle noise model contains gamma distributed multiplicative part which is dominant and Gaussian distributed additive part which is in low amount and mostly neglected in literature. In this study, a novel sparsity-driven speckle reduction method is proposed that takes both multiplicative noise model and additive noise model into consideration. The proposed speckle reduction method uses a cost function with multiplicative and additive data terms besides the total variation smoothness term. Also, an efficient and stable numerical minimization scheme is proposed for the proposed cost function that deals with multiplicative and additive noise. Speckle reduction performance of the proposed method is shown on synthetically generated SAR images and real-world SAR images.

Parallel techniques for speckle noise reduction in medical ultrasound images

Parallel computing is used to accelerate the computations in the reduction of speckle noise in medical ultrasound images. A new hybrid algorithm for speckle noise reduction in medical ultrasound images is proposed and compared to an existing efficient method. The method combines in an efficient manner the advantages of several denoising filters using local and non-local information. A comparative study is carried out using quantitative and qualitative measures and showing the competitiveness of the proposed method. Both methods are parallelized using OpenMP, and a hybrid combination of MPI and OpenMP. The parallel implementations are compared to the sequential ones, obtaining significative values of speedup.

Speckle removal in variable density ESPI fringe patterns with TGV–Hilbert–Shearlet algorithm

The fringe pattern obtained by optical interference is susceptible to speckle noise, which causes the fringes to be blurred to some extent. It is a critical step to achieve fringe smoothness while removing speckle noise. In this paper, we propose a new variable image decomposition model TGV–Hilbert–Shearlet, in which low-density fringes, high-density fringes, and noise are described by the total generalized variation (TGV) space, adaptive Hilbert space, and shearlet space, respectively. By assigning appropriate parameters, the image noise reduction and the fringe structural smoothness can be achieved optimally. We test the proposed method on five computer simulation ESPI fringe patterns and three experimentally obtained ESPI images. In addition, we compare it with BL-Hilbert-L2 and Window Fourier Filter (WFF), which were proved to be effective in denoising. On the basis of regions marked in different images, the signal-to-noise ratio and the average equivalent number of looks are calculated to better characterize the denoising effect and fringe smoothness. Vast experiments show that the proposed TGV–Hilbert–Shearlet method can effectively reduce speckle noise in ESPI images, protect fringe structural information and improve image quality in all aspects compared with BL-Hilbert-L2 and WFF methods.

AN APPROACH FOR IMPROVING ACCURACY OF CHANGE DETECTION IN MULTI-TEMOPRAL SAR IMAGES

Algorithms of change detection in multi-temporal SAR images have received great interests for recent decades, and been widely applied in natural resource supervision activities. However, these algorithms still expose the limitation of detection accuracy due to inhenrent presence of speckle noise in SAR images. This paper developed a novel approach of change detection in multi-temporal SAR images of sea surface. The algorithm has increased accuracy of change detection in multi-temporal SAR images of sea surface compared with recent other methods.

Review on automated follicle identification for polycystic ovarian syndrome

Polycystic Ovarian Syndrome (PCOS), is a condition of the ovary consisting numerous follicles. Accurate size and number of follicles detected are crucial for treatment. Hence the diagnosis of this condition is by measuring and calculating the size and number of follicles existed in the ovary. For diagnosis, ultrasound imaging has become an effective tool as it is non-invasive, inexpensive and portable. However, the presence of speckle noise in ultrasound imaging has caused an obstruction for manual diagnosis which are high time consumption and often produce errors. Thus, image segmentation for ultrasound imaging is critical to identify follicles for PCOS diagnosis and proper health treatment. This paper presents different methods proposed and applied in automated follicle identification for PCOS diagnosis by previous researchers. In this paper, the methods and performance evaluation are identified and compared. Finally, this paper also provided suggestions in developing methods for future research.

Polarimetric SAR image despeckling using bandelet transform based on additive–multiplicative speckle model

Efficient speckle filtering algorithms are required for the effective use of polarimetric synthetic aperture radar (SAR) technology in remote sensing and surveillance applications. Nevertheless many techniques have been proposed over the past two decades to decrease the speckle noise in polarimetric SAR images, they are all based on the multiplicative speckle noise model. In order to fully utilise the advantages of polarimetry of these images, an additive–multiplicative noise model is explored. Coupled with this, bandelet based Bayesian thresholding is used to tap the advantages of transform domain filtering. Here the elements of the covariance matrix are processed differently for diagonal and off-diagonal elements to achieve maximum benefits. The proposed filtering scheme is evaluated using airborne and spaceborne polarimetric images and compared against state-of-the-art techniques. Results indicate that the proposed method reduces the speckle content while retaining the geometrical features in these images.

Removing Speckle Noise in Synthetic Aperture Radar Images using Linear Filters

Speckle is a granular aggravation, typically demonstrated as a multiplicative noise that influences Synthetic aperture radar (SAR) pictures, just as every single cognizant picture. In the course of the most recent three decades, a few techniques have been proposed for the decrease of spot, or despeckling, in SAR pictures. The examination begins with the linear filtering, non-linear filtering, adaptive filtering, and hybrid filtering. In spite of the fact that the old style straight separating strategies have lower execution similarly, the hybridization between them beats than the as of late proposed techniques. Be that as it may, the cautious determination of such filters and their impacting request exceptionally influences the presentation of such filtering methodologies. In this paper, Hybrid filtering is proposed for SAR despeckling, which involves the improved variants of frost, median and mean filters. The presentation of the proposed framework is broke down and contrasted and the as of late SAR despeckling strategies. The outcomes demonstrate that the hybrid filters are focused as they can despeckle the SAR pictures superior to the current methods

Optical coherence tomography image denoising using a generative adversarial network with speckle modulation.

Optical coherence tomography (OCT) is widely used for biomedical imaging and clinical diagnosis. However, speckle noise is a key factor affecting OCT image quality. Here, we developed a custom generative adversarial network (GAN) to denoise OCT images. A speckle-modulating OCT (SM-OCT) was built to generate low speckle images to be used as the ground truth. In total, 210 000 SM-OCT images were used for training and validating the neural network model, which we call SM-GAN. The performance of the SM-GAN method was further demonstrated using online benchmark retinal images, 3D OCT images acquired from human fingers and OCT videos of a beating fruit fly heart. The denoise performance of the SM-GAN model was compared to traditional OCT denoising methods and other state-of-the-art deep learning based denoise networks. We conclude that the SM-GAN model presented here can effectively reduce speckle noise in OCT images and videos while maintaining spatial and temporal resolutions.

Speckle Noise Reduction in Ultrasound Images for Improving the Metrological Evaluation of Biomedical Applications: An Overview

In recent years, many studies have examined filters for eliminating or reducing speckle noise, which is inherent to ultrasound images, in order to improve the metrological evaluation of their biomedical applications. In the case of medical ultrasound images, said noise can produce uncertainty in the diagnosis because details, such as limits and edges, should be preserved. Most algorithms can eliminate speckle noise, but they do not consider the conservation of these details. This paper describes, in detail, 27 techniques that mainly focus on the smoothing or elimination of speckle noise in medical ultrasound images. The aim of this study is to highlight the importance of improving said smoothing and elimination, which are directly related to several processes (such as the detection of regions of interest) described in other articles examined in this study. Furthermore, the description of this collection of techniques facilitates the implementation of evaluations and research with a more specific scope. This study initially covers several classical methods, such as spatial filtering, diffusion filtering, and wavelet filtering. Subsequently, it describes recent techniques in the field of machine learning focused on deep learning, which are not yet well known but greatly relevant, along with some modern and hybrid models in the field of speckle-noise filtering. Finally, five Full-Reference (FR) distortion metrics, common in filter evaluation processes, are detailed along with a compensation methodology between FR and Non-Reference (NR) metrics, which can generate greater certainty in the classification of the filters by considering the information of their behavior in terms of perceptual quality provided by NR metrics.

基于OCT成像信号振幅差分概率密度分布的图像散斑噪声抑制方法

基于OCT成像信号振幅差分概率密度分布的图像散斑噪声抑制方法

Selective retinex enhancement based on the clustering algorithm and block-matching 3D for optical coherence tomography images.

It is important to enhance the contrast and remove the speckle noise for optical coherence tomography (OCT) images. In this paper, we propose a selective retinex enhancement method based on the fuzzy c-means (FCM) clustering algorithm to enhance only the structure part in OCT images and combines with the block-matching 3D (BM3D) algorithm for filtering. In the proposed selective retinex enhancement method, we first calculate the feature image of the original image, which includes the mean value and standard deviation of each pixel in the original image and its correlation image. Second, by applying the FCM clustering algorithm to the feature image, a mask is generated that can distinguish the structure part from the background part in the OCT image. Then, the mask is applied to the multi-scale retinex algorithm, and only the structure part in the OCT image is enhanced. Moreover, the BM3D method is applied to filter the enhanced image. Experimental results demonstrate that the proposed method performs impressively in improving the contrast and removing the speckle noise of OCT images, and it provides better quantitative performance in terms of signal-to-noise ratio, contrast-to-noise ratio, equivalent number of looks, and the edge preservation parameter $ \beta $β.

IVUS Image Segmentation Using Superpixel-Wise Fuzzy Clustering and Level Set Evolution

Reliable detection of the media-adventitia border (MAB) and the lumen-intima border (LIB) in intravascular ultrasound (IVUS) images remains a challenging task that is of high clinical interest. In this paper, we propose a superpixel-wise fuzzy clustering technique modified by edges, followed by level set evolution (SFCME-LSE), for automatic border extraction in 40 MHz IVUS images. The contributions are three-fold. First, the usage of superpixels suppresses the influence of speckle noise in ultrasound images on the clustering results. Second, we propose a region of interest (ROI) assignment scheme to prevent the segmentation from being distracted by pathological structures and artifacts. Finally, the contour is converged towards the target boundary through LSE with an appropriately improved edge indicator. Quantitative evaluations on two IVUS datasets by the Jaccard measure (JM), the percentage of area difference (PAD), and the Hausdorff distance (HD) demonstrate the effectiveness of the proposed SFCME-LSE method. SFCME-LSE achieves the minimal HD of 1.20 ± 0.66 mm and 1.18 ± 0.70 mm for the MAB and LIB, respectively, among several state-of-the-art methods on a publicly available dataset.

A Hybrid Neural Filter (HNF) Based on Adaptive Median and Weiner Techniques for Reducing Speckle Noise of Ultrasound Liver Tumor Images

Pre-processing of image is considered to be an important aspect medical image analysis in order to enhance its quality. Noise reduction is performed for improvement in the visual quality and removing redundant image values. Ways of pre-processing of image are a necessity, for removing the noise and for quality enhancement of the image. Before applying any approach on medical images, measures used to pre-process seem to be vital for limiting the abnormalities’ findings with no influence from background of the medical image. In this work, a filter is proposed based on Adaptive Median and Weiner Hybrid Neural Filter for noise reduction. The review of filtering techniques is implemented using MATLAB platform, followed by comparison of results with different filtering techniques to show the system effectiveness.

Segmentation of sonar images with intensity inhomogeneity based on improved MRF

Sonar is one of the most important tools for underwater object detection and submarine topography reconstruction. To classify sonar images automatically and accurately is essential for the navigation and path planning of autonomous underwater vehicles (AUV). However, for the intensity inhomogeneity and speckle noise in sonar images, it is difficult to obtain segmentation results of high accurate rate. To address these issues, in this paper, we advocate a segmentation method incorporating simple linear iterative clustering (SLIC) and adaptive intensity constraint into Markov random field (MRF), to segment sonar images with intensity inhomogeneity into the object highlight, the object shadow and the background areas. The main procedures of the proposed work are as follows: first, SLIC is used to separate sonar images into homogeneous super pixels, and second the homogeneity patches, with a novel intensity constraint strategy, is utilized to optimize the segmentation result of MRF at each iteration. Experimental results reveal that the proposed method performs well and fast on real sonar images which have intensity inhomogeneity problem.

Speckle denoising in optical coherence tomography images using residual deep convolutional neural network

Optical Coherence Tomography (OCT) is an emerging imaging modality used for diagnosis of ocular diseases like age-related macular degeneration (AMD) and macular edema. OCT imaging is a non-invasive technique to capture cross-sectional volumes of the retinal areas of human eye. Due to coherent nature of image acquisition process, OCT images suffer from granular multiplicative speckle noise. Presence of speckle noise in OCT images makes its clinical analysis difficult for the experts. The same is the problem with the development of computer aided diagnosis (CAD) systems for detection of ocular diseases. Speckle noise is granular in nature and interferes with the diagnostic observations made using OCT images and the segmentation of different OCT layers. This work presents an efficient OCT denoising technique using residual convolutional neural network. The proposed technique will not only help experts in analysis of OCT images, but can also act as first step to construct CAD systems for ocular diseases. The performance of the proposed approach is evaluated on Duke (SD-OCT) and Topcon (3D-OCT) image databases based on visual and parametric observations. The performance of the proposed method on parameters like PSNR, SSIM, MSR, CNR, and ENL is compared with the state-of-the-art speckle denoising methods. It is observed that the proposed approach performs better as compared to the methods referred from literature on both visual and parametric evaluations.

Improved FAST corner‐detection method

An improved features from accelerated segment test (FAST) corner-detection method for suppressing SAR image speckle noise is proposed here. The direction of the corner candidate is first computed by the law of universal gravitation and is introduced into corner detection to divide the 16 pixels around the corner candidate into two groups without repeatedly searching. Then, the similarity between the corner candidate and its surrounding pixels is computed to suppress the points affected by SAR speckle noise. Finally, the phenomenon of corner density is eliminated based on non-maximum suppression. The experimental results show that authors’ method has better suppressing effect compared with FAST-8 and SIFT corner-detection methods.

BM3D-based total variation algorithm for speckle removal with structure-preserving in OCT images.

In this paper, we propose a total variation based on block matching 3D (BM3D-TV method), which includes the total variation regular term, the data fidelity term, and the block matching term. In addition, we also propose a fast numerical algorithm based on the split Bregman iteration for the proposed method. By assigning suitable weights to the data fidelity term and block matching term, the image noise reduction and the image structural characteristics can be matched optimally. We test the proposed method on six human retinal and one mouse skin optical coherence tomography (OCT) images respectively, and also compare it with total variation (TV) and BM3D, which were proved to be effective in denoising. The performances of these methods are quantitatively evaluated in terms of the signal-to-noise ratio, the contrast-to-noise ratio, and the averaged equivalent number of homogeneous areas at the aspects of speckle reduction and structure protection. Vast experiments show that the BM3D-TV method can effectively reduce speckle noise in OCT images, protect important structural information and improve image quality, compared with the BM3D and TV methods.

De-Speckling of Ultrasound Images Using Local Statistics-Based Trilateral Filter

Speckle noise in ultrasound images is a major hindrance for the automation of segmentation, detection, classification and measurements of region of interest, to assist clinician for diagnosing pathologies. Speckle noise occurs due to constructive and destructive interference of the echo signals reflected from the target and has a granular appearance. Various techniques have been devised for speckle reduction. Most of these techniques are based on adaptive filters, wavelet transform and anisotropic diffusion filters. In this paper, a new speckle reduction technique based on the trilateral filter and local statistics of the image has been developed. The local speckle content of the image influences the trilateral filtering. The trilateral filter is a robust edge preserving filter which considers the similarity of neighboring regions in terms of adjacency, intensity and edge details. Hence, the new method preserves the finer details of the ultrasound images in the process of filtering speckle noise. The proposed technique is validated using synthetic, simulated and real-time clinical ultrasound images. Comparison of the proposed technique with the existing speckle removal algorithms in terms of quality metrics such as MSE, PSNR, UQI, SSI, FoM has been made and best results are obtained for the proposed technique.

SAR Image Registration Using an Improved SAR-SIFT Algorithm and Delaunay-Triangulation-Based Local Matching

Synthetic aperture radar (SAR) image registration is a challenging task in remote sensing because of the presence of significant intensity as well as geometric differences between the images. Moreover, the influence of multiplicative speckle noise in SAR images adversely affects the registration result. The synthetic aperture radar-scale invariant feature transform (SAR-SIFT) algorithm, specially developed for the SAR images, has been used popularly for the automatic registration of SAR images based on global feature matching. However, the algorithm suffers from lack of controllability of the number of extracted features and uneven distribution of the features. Moreover, global matching of the features reduces the number of correct matches. To address these problems, we propose an improved SAR-SIFT (I-SAR-SIFT) algorithm, which extracts features from the SAR images by considering three important factors of feature-based image registration: stability, distinctiveness, and distribution. Moreover, a novel approach is adapted to improve the distinctiveness of the feature–descriptor in the presence of speckle noise. In addition, a Delaunay-triangulation-based local matching algorithm is introduced to significantly improve the matching performance in SAR images. Experiments on different pairs of SAR images show that the proposed method gives better performance than the existing SAR image registration methods.