Denoising Filter Research Articles

Efficient digit serial architecture for sign based least mean square adaptive filter for denoising of artefacts in ECG signals

Variants of Least Mean Square algorithm, namely the Sign Error, Sign Regressor and Sign-Sign Least Mean Square algorithms are meant for reducing computational complexity in digital adaptive filters in denoising electrocardiogram signals. In adaptive filtering using Sign-Sign algorithm, the adders and multipliers in the existing method take the coefficients in canonic signed power of two form which assures that between two non-zero there should be one zero in signed power of two digits and they can process only one bit per clock cycle and so is the filter resulting in low speed. In this paper, digit serial adder and digit serial multiplier are proposed to design the Sign-Sign Least Mean Square filter, which increases the speed by processing two bits per clock cycle. The digit serial adder and digit serial multiplier are used to design the Sign-Sign based adaptive filter in which unfolding technique is used for digit serial realisation. The unfolding factor of K = 2 is chosen to increase the speed of the design. The coefficients and its inputs are represented in Canonic Signed Digit form. Adder, multiplier, adaptive filter of both existing and proposed method are simulated using Mentor Graphics and synthesised using Synopsys Design Compiler Tool with 90 nm technology. From the results it is observed that the speed of the proposed digit serial Sign-Sign Least Mean Square filter is increased by 54.5% and the area of the filter is decreased by 4.21% compared to the bit serial Sign-Sign Least Mean Square filter.

A Density Current Modeled Adaptive Weighted Average Despeckling Filter for Ultrasound Thyroid Images

Speckle noise reduction is an important preprocessing stage for ultrasound medical image processing. In this paper, an adaptive weighted average despeckling algorithm is proposed based on density currents equation. An analogy between the gravity current density model for sea breeze and speckle noise affected image is used and a denoising filter is designed. Suitable threshold values to distinguish between different regions in the image are selected for the denoising filter by considering the structural similarity and mean square error parameters of the particular image. Hence the proposed denoising filter adaptively changes the threshold values depending on nature of image and magnitude of noise in the image. For each pixel, weights have been changed adaptively by considering local neighbours in different directions. Experimental results of proposed method for natural images, Field II simulated images and real ultrasound images, show that the proposed method outperforms other existing methods in terms of signal to noise ratio, Structural similarity index and Ultrasound despeckling assessment index. It is also shown that proposed filter is able to improve the visual quality of Ultrasound images and better preserves image structural details compared to existing methods.

Efficient spectrum sensing technique based on energy detector, compressive sensing, and de-noising techniques

Highdata rate cognitive radio (CR) systems require high speed Analog-to-Digital Converters (ADC). This requirement imposes many restrictions on the realization of the CR systems. The necessity of high sampling rate can be significantly alleviated by utilizing analog to information converter (AIC). AIC is inspired by the recent theory of Compressive Sensing (CS), which states that a discrete signal has a sparse representation in some dictionary, which can be recovered from a small number of linear projections of that signal. This paper proposes an efficient spectrum sensing technique based on energy detection, compression sensing, and de-noising techniques. De-noising filters are utilized to enhance the traditional Energy Detector performance through Signal-to-Noise (SNR) boosting. On the other hand, the ordinary sampling provides an ideal performance at a given conditions. A near optimal performance can be achieved by applying compression sensing. Compression sensing allows signal to be sampled at sampling rates much lower than the Nyquist rate. The system performance and ADC speed can be easily controlled by adjusting the compression ratio. In addition, a proposed energy detector technique is introduced by using an optimum compression ratio. The optimum compression ratio is determined using a Genetic Algorithm (GA) optimization tool. Simulation results revealed that the proposed techniques enhanced system performance.

Nonparametric de-noising filter optimization using structure-based microscopic image classification.

The Local Polynomial Approximation (LPA) is a nonparametric filter that performs pixel-wise polynomial fit on a certain neighborhood. This filter can be supported by the Intersection of Confidence Interval rule (ICI) as an adaptation algorithm to identify the most suited neighborhood at which the polynomial assumptions provide superior fit for the observations. However, the LPA-ICI is considered to be a near-optimal de-noising filter. Moreover, the ICI rule has several parameters that affect its performance. The current study applied an optimization algorithm, namely the Particle swarm optimization (PSO) to determine the optimal ICI parameter values for microscopic images de-noising. As the ICI parameters are image's structure based, bag-of-features classifier is used to classify the images based on their structure into different classes. Afterward, a generated optimal ICI parameters' table was created using the LPA-ICI-PSO for further direct use without optimization. This table included the optimal ICI parameters based on the image structure. Based on the image category, the generated table can be used to attain the suitable ICI optimal parameters without using PSO. This guarantees less computational time along with the optimal de-noising compared to the LPA-ICI as established by the performance metrics. The experimental results established the superiority of the proposed LPA-ICI-PSO over the classical LPA-ICI filter.

A new approach of weighted gradient filter for denoising of medical images in the presence of Poisson noise

A new approach of weighted gradient filter for denoising of medical images in the presence of Poisson noise

Performance enhancement of MPM using SNR boosting techniques

Signal to noise ratio (SNR) boosting is one of the most important research areas in signal processing. The effectiveness of SNR boosting is not limited to a specific application rather, it is widely used in image processing, signal processing, cognitive radio, MIMO systems, digital beam forming, and direction of arrival (DOA) estimation …etc. In this paper, the recursive least square (RLS) and wavelet based de-noising filters are exploited for SNR boosting in DOA estimation techniques for further performance enhancement. The matrix pencil method (MPM) as an effortlessness and high resolution DOA estimation technique is taken as a test case. That is because it suffers from performance deterioration under low SNR regimes. The simulation results reveal that the MPM based RLS de-noising filter outperforms the MPM based wavelet de-noising filter and the traditional MPM in terms of mean square error (MSE) especially at low SNR regimes.

Fractional order integration and fuzzy logic based filter for denoising of echocardiographic image

BackgroundUltrasound is widely used for imaging due to its cost effectiveness and safety feature. However, ultrasound images are inherently corrupted with speckle noise which severely affects the quality of these images and create difficulty for physicians in diagnosis. To get maximum benefit from ultrasound imaging, image denoising is an essential requirement. MethodTo perform image denoising, a two stage methodology using fuzzy weighted mean and fractional integration filter has been proposed in this research work. In stage-1, image pixels are processed by applying a 3 × 3 window around each pixel and fuzzy logic is used to assign weights to the pixels in each window, replacing central pixel of the window with weighted mean of all neighboring pixels present in the same window. Noise suppression is achieved by assigning weights to the pixels while preserving edges and other important features of an image. In stage-2, the resultant image is further improved by fractional order integration filter. ResultsEffectiveness of the proposed methodology has been analyzed for standard test images artificially corrupted with speckle noise and real ultrasound B-mode images. Results of the proposed technique have been compared with different state-of-the-art techniques including Lsmv, Wiener, Geometric filter, Bilateral, Non-local means, Wavelet, Perona et al., Total variation (TV), Global Adaptive Fractional Integral Algorithm (GAFIA) and Improved Fractional Order Differential (IFD) model. Comparison has been done on quantitative and qualitative basis. For quantitative analysis different metrics like Peak Signal to Noise Ratio (PSNR), Speckle Suppression Index (SSI), Structural Similarity (SSIM), Edge Preservation Index (β) and Correlation Coefficient (ρ) have been used. Simulations have been done using Matlab. Simulation results of artificially corrupted standard test images and two real Echocardiographic images reveal that the proposed method outperforms existing image denoising techniques reported in the literature. ConclusionThe proposed method for denoising of Echocardiographic images is effective in noise suppression/removal. It not only removes noise from an image but also preserves edges and other important structure.

Application of Improved SVM Algorithm in Color Image De-Noising

It cannot avoid the noise interference in image processing, whether it is image generation, or image transmission, among them, the most typical noise is salt and pepper noise and Gaussian noise. The salt and pepper noise will cause the image showing the random distribution of noise points, thus greatly reduce the image quality. The Gaussian noise affects the input, collection and output of the image processing. Gaussian noise will make the image blurred. Therefore, the image de-noising plays a very important role in image processing. It has direct influence on image segmentation, feature extraction and image recognition. As is known to all, the support vector machine has the advantages of solving the problem of nonlinear, high dimension and local minimum points. In this article, we use this advantage to propose an image de-noising method which is based on it. The method uses support vector regression to construct the filter for image de-noising. The feature extraction and training samples are designed to suppress different types of noise. Firstly, we use the noise pixel as the center of the 5*5 window, and generate the input vector of SVM from row to column. Secondly, we set the output of the support vector filter as an image that is not contaminated by noise. At this point, we get the training samples of SVM filter. In addition, the parameter selection of support vector machine has a great influence on the result of image de-noising. Therefore, the particle swarm optimization algorithm is proposed in this article to optimize the parameters of SVM. Finally, we adding the simulated salt and pepper noise and Gaussian noise in the original Lena image, and using several methods to carry out the de-noising experiment. From the experimental results we can see that the de-noising effect of filtering algorithm of this paper is very good for the two kinds of noise. It can effectively remove the noise, and better maintain the details and the color of the image.

A two-stage framework for denoising electrooculography signals

Abstract Denoising of electrooculography (EOG) signals is a challenging task as the noise and signal share the same frequency band. This paper proposes a two-stage framework for denoising EOG signals. The first stage approach is based on preserving the nature of eye movements while the second stage is based on the nature of noise (Gaussian or not). In the first stage, denoising is carried out using one out of four filtering methods, each filter being optimal for a particular EOG pattern. The four methods used in the first stage are linear bandpass filtering, stationary wavelet transform (SWT), empirical mode decomposition (EMD) and median filtering. The Stage I framework selects the output that provides the highest estimated signal to noise ratio (SNR). In case, the Stage I filtering does not provide a significant SNR, the system uses Stage II filtering. In the second stage, we use two recursive state estimators, i.e. a Kalman filter and a particle filter for further denoising. The two-stage method is found to provide a better SNR as compared to a single stage method.

Application of reversible denoising and lifting steps with step skipping to color space transforms for improved lossless compression

Reversible denoising and lifting steps (RDLS) are lifting steps integrated with denoising filters in such a way that, despite the inherently irreversible nature of denoising, they are perfectly reversible. We investigated the application of RDLS to reversible color space transforms: RCT, YCoCg-R, RDgDb, and LDgEb. In order to improve RDLS effects, we propose a heuristic for image-adaptive denoising filter selection, a fast estimator of the compressed image bitrate, and a special filter that may result in skipping of the steps. We analyzed the properties of the presented methods, paying special attention to their usefulness from a practical standpoint. For a diverse image test-set and lossless JPEG-LS, JPEG 2000, and JPEG XR algorithms, RDLS improves the bitrates of all the examined transforms. The most interesting results were obtained for an estimation-based heuristic filter selection out of a set of seven filters; the cost of this variant was similar to or lower than the transform cost, and it improved the average lossless JPEG 2000 bitrates by 2.65% for RDgDb and by over 1% for other transforms; bitrates of certain images were improved to a significantly greater extent.

Variance Stabilization for Noisy+Estimate Combination in Iterative Poisson Denoising

We denoise Poisson images with an iterative algorithm that progressively improves the effectiveness of variance-stabilizing transformations (VST) for Gaussian denoising filters. At each iteration, a combination of the Poisson observations with the denoised estimate from the previous iteration is treated as scaled Poisson data and filtered through a VST scheme. Due to the slight mismatch between a true scaled Poisson distribution and this combination, a special exact unbiased inverse is designed. We present an implementation of this approach based on the BM3D Gaussian denoising filter. With a computational cost at worst twice that of the noniterative scheme, the proposed algorithm provides significantly better quality, particularly at low signal-to-noise ratio, outperforming much costlier state-of-the-art alternatives.

A Wireless BCI and BMI System for Wearable Robots

To increase the performance of a brain–computer interface and brain–machine interface system, we propose some methods and algorithms for electroencephalograph (EEG) signal analysis. The recorded EEG signal is transmitted to the computer and the upper limb robotic arm interface via a bluetooth. To obtain effective commands from brain, the recorded EEG signal is processed by a front filter, denoise filter, feature extraction, and classification, while the personal computer software and upper limb arm are driven by EEG-based commands. Through the encoders and gyroscopes on the upper limb arm, we can acquire some feedback signals in real time, such as joint angle, arm accelerated speed, and angular speed. The theory of wavelet denoising method, common spatial pattern algorithm and linear discriminant analysis algorithm are investigated in this paper. The simulations and experiments demonstrate the effectiveness and accuracy of these algorithms on EEG signal denoising, feature extraction, and classification.

Performance enhancement of MPM DoA estimation technique using wavelet De-noising filter

Direction-of-arrival (DoA) estimation is now an imperative part in many radar applications and localization techniques. There are numerous algorithms that have been studied in the previous decades for DoA, for example: MUSIC, ESPRIT, and Matrix Pencil Method (MPM), which are subspace super resolution methods. MPM is one of the most commonly used subspace based techniques. It is generally utilized for DoA estimation because of its effortlessness and high resolution contrasted with other subspace techniques. But, it suffers from performance deterioration under low Signal-to-Noise Ratio (SNR) conditions. This paper, explores the possibility of utilizing the wavelet de-noising technique to intercept the degradation in the performance of MPM under different SNR values. Wavelet De-noising is intended to remove noise or distortion from signals while retaining the original quality of the signal. The simulation results indicate that the Daubechies wavelet (db12) at 5 levels of decomposition is the most suitable wavelet for de-noising the signals under test. Also, the results show that the proposed wavelet de-noising matrix pencil method (WDMPM) outperforms the traditional MPM.Performance Enhancement of MPM DoA Estimation Technique using Wavelet De-noising Filter

Sequence-to-Sequence Similarity-Based Filter for Image Denoising

Image denoising has been a well-studied problem for imaging systems, especially imaging sensors. Despite remarkable progress in the quality of denoising algorithms, persistent challenges remain for a wide class of general images. In this paper, we present a new concept of sequence-to-sequence similarity (SSS). This similarity measure is an efficient method to evaluate the content similarity for images, especially for edge information. The approach differs from the traditional image processing techniques, which rely on pixel and block similarity. Based on this new concept, we introduce a new SSS-based filter for image denoising. The new SSS-based filter utilizes the edge information in the corrupted image to address image denoising problems. We demonstrate the filter by incorporating it into a new SSS-based image denoising algorithm to remove Gaussian noise. Experiments are performed over synthetic and experimental data. The performance of our methodology is experimentally verified on a variety of images and Gaussian noise levels. The results demonstrate that the proposed method’s performance exceeds several current state-of-the-art works, which are evaluated both visually and quantitatively. The presented framework opens up new perspectives in the use of SSS methodologies for image processing applications to replace the traditional pixel-to-pixel similarity or block-to-block similarity.

Scattering-Based Nonlocal Means SAR Despeckling

Speckle noise greatly limits both synthetic aperture radar (SAR) data human readability, especially for non-SAR-expert users, and performance of automatic processing and information retrieval procedures by computer programs. Therefore, despeckling of SAR images is an essential preprocessing step in SAR data analysis, processing, and modeling, as well as in information retrieval and inversion procedures. Up to now, one of the most accurate and promising despeckling approaches—among those based on a single SAR image—is the one relying on the nonlocal means concepts. However, at the best of our knowledge, most of the state of the art considers the despeckling problem only within a statistical framework, completely discarding the electromagnetic phenomena behind SAR imagery formation. In this paper, we introduce the novel idea of a physical-based despeckling, taking into account meaningful physical characteristics of the imaged scenes. This idea is realized via the implementation of a physical-oriented probabilistic patch-based (PPB) filter based on a priori knowledge of the underlying topography and analytical scattering models. This filter is suitable for SAR images of natural scenes presenting a significant topography. An adaptive version of the proposed scattering-based PPB filter for denoising of SAR images including both mountainous and flat areas is also developed. The performances of the proposed filter and its adaptive version are evaluated both qualitatively and quantitatively in numerical experiments using both simulated and actual SAR images. The proposed technique exhibits performance superior w.r.t. the standard PPB filter and comparable or, in some cases, superior to the state of the art, both in terms of speckle reduction and texture and detail preservation.

An Anisotropic Convection-Diffusion Model Using Tailored Finite Point Method for Image Denoising and Compression

AbstractIn this paper we consider an anisotropic convection-diffusion (ACD) filter for image denoising and compression simultaneously. The ACD filter is discretized by a tailored finite point method (TFPM), which can tailor some particular properties of the image in an irregular grid structure. A quadtree structure is implemented for the storage in multi-levels for the compression. We compare the performance of the proposed scheme with several well-known filters. The numerical results show that the proposed method is effective for removing a mixture of white Gaussian and salt-and-pepper noises.

Real-Time Neutron Imaging to Detect Origin of Blocking in Drug Injection Devices.

Neutron imaging is a nondestructive method based on radiography using neutrons and is suitable for detecting small amounts of aqueous liquids even in metallic casing/sheath/tubing. This property has here been used to visualize the distribution of a drug product in a syringe needle of a drug injection device. In the static case the method clearly showed the difference between needles that were empty, full, or contained a mix of gas and liquid. A second investigation was aimed to visualize the dynamic behavior of an auto-injector device. In this experiment the imaging method was capable of following the injection phase of the device. By analyzing the acquired images in time and space it is possible to measure the injection velocity curves of the piston and drug, respectively.

Image modeling based on a 2-D stochastic subspace system identification algorithm

Fitting a causal dynamic model to an image is a fundamental problem in image processing, pattern recognition, and computer vision. In image restoration, for instance, the goal is to recover an estimate of the true image, preferably in the form of a parametric model, given an image that has been degraded by a combination of blur and additive white Gaussian noise. In texture analysis, on the other hand, a model of a particular texture image can serve as a tool for simulating texture patterns. Finally, in image enhancement one computes a model of the true image and the residuals between the image and the modeled image can be interpreted as the result of applying a de-noising filter. There are numerous other applications within the field of image processing that require a causal dynamic model. Such is the case in scene analysis, machined parts inspection, and biometric analysis, to name only a few. There are many types of causal dynamic models that have been proposed in the literature, among which the autoregressive moving average and state-space models (i.e., Kalman filter) are the most commonly used. In this paper we introduce a 2-D stochastic state-space system identification algorithm for fitting a quarter plane causal dynamic Roesser model to an image. The algorithm constructs a causal, recursive, and separable-in-denominator 2-D Kalman filter model. The algorithm is tested with three real images and the quality of the estimated images are assessed.

Computational Tools to Improve Visualization by Cryo-Electron Tomography

Cryo-electron tomography (CryoET) can directly visualize frozen, hydrated macromolecules inside cells without staining artifacts or fluorescent tags at nanometer resolution. However, imaging tilted specimens is limited to ±60° tilt, causing a “missing wedge” of data that induces smearing artifacts in tomograms. Furthermore, each image is subject to a defocusing artifact known as the contrast transfer function (CTF) of the electron microscope. Lastly, high noise and low contrast make feature identification difficult and error prone. Although contrast is typically improved by denoising (filtering), the tomographic reconstruction process introduces additional artifacts that can be accentuated when filters are applied to tomograms in 3-D.We developed tools for CTF correction in cryoET tiltseries and to fill in the missing wedge. We implemented them in EMAN2, a widely used software package for cryo elctron microscopy data processing. Using our first too, we applied CTF correction on cryoET tiltseries of viral particles of known structure recorded with a direct electron detector (DE20). Images collected in “movie-mode” were first corrected for drift. We achieve a resolution improvement from ∼25A to ∼13A. Our second tool “fills in” the missing wedge with reprojections of tomographic data from angular directions not accessible experimentally in the electron microscope. Results on viral particle data show decreased missing wedge artifacts (smearing). We also applied pre-reconstruction denoising filters directly to 2-D images in tiltseries, which improved the accuracy of automated particle picking (measured against manual particle picking, considered ground truth).Lastly, we applied pre-reconstruction denoising filters to tiltseries of mammalian cells to minimize artifacts and enhance the reliability of feature identification. For example, in platelets, microtubules and the cell membrane become clearer with pre-filtration, compared to the standard post-reconstruction filtering approach, and mitochondria become more easily distinguishable from granules.

Linear Filtering and Total Variation De-noising of Composite Noises: Different Architectures and Simulation Study

Combined techniques consisting of traditional linear filtering and total variation de-noising are studied to de-noise images polluted with composite noises from different sources. Series and parallel approaches are compared theoretically. Some further discussions are given about parameters and their effect on the optimization problem. A simultaneous method is proposed and simulated using the convex optimization toolbox, comparing different alternative methods with respect to signal to noise ratio and error covariance matrix. The proposed method could be used to enhance the smoothing performance.