Adaptive Threshold and Directional Weighted Median Filter-Based Impulse Noise Removal Method for Images

In this paper an effort has been made to device an algorithm for highly corrupted images. In this paper a novel Modified Weighted Based(MWB) filter has been proposed, which is based on the weighted differences between the current pixel and its neighbors aligned with four main directions. Simulations showed that the MWB filter provides optimal performances of suppressing impulse with high noise level which may enhance the performance in terms of removal of random-valued impulse noise compared to the directional weighted median (DWM) filter along with other filters like signal-dependent rank order mean (SDROM)filter, multistate median (MSM) filter, adaptive center weighted median (ACWM) filter, pixel-wise MAD(PWMAD) filter etc.

Most of the impulse denoisers are either median filter‐based or fuzzy filter‐based, which can only perform well in low noise conditions. This study presents an efficient convolutional neural network (CNN) with particle swarm optimisation (PSO) model for high‐density impulse noise removal. The proposed high‐density impulse noise detection and removal model mainly consists of two parts: the impulse noise removal and impulse noisy pixel detection for restoration. The authors’ model initially leverages the powerful ability of deep CNN architecture to separate noise from the noisy image, then adopts PSO to pinpoint the most optimised threshold values for detecting impulse noisy pixels. An ensemble of these algorithms is an intelligent and adaptive solution, producing a clean output while preserving significant pixel information. Targeting to solve high‐density impulse noise problems, the authors have trained their model with a massive collection of natural images and 14 standard testing images are used for validation purposes. In order to validate the robustness of the proposed method, different levels of high‐density impulse noise are considered. Based on the final denoised images, their model has proven its reliability, in terms of both visual quality and quantitative evaluation, on greyscale and colour images.

The objective of this paper was to analyze several approaches to the probabilistic modeling of Hokuyo-type laser rangefinder (LRF). These models, which reflect mostly the mixed pixels, beam width, signal, and impulse noise, were compared mutually. Motivation for creating these models was the failure of existing LRF models to faithfully represent the features of sensed environment. Moreover, the characteristics of each particular model were highlighted, and their suitability for various applications was determined. This paper also dealt with data fusion by selected model and its representation in the occupancy grid. This was based on the classical Bayes’ rule, which was adapted for the desired characteristics of data fusion. Finally, in order to improve data from the sensor, data pre-processing was proposed. Mentioned data pre-processing was proposed with the emphasis on the elimination of impulsive noise. The overall concept of data processing from the LRF can be regarded as versatile and robust.

The elimination of noise from images becomes a trending field in image processing. Imagesmay got corrupted by random change in pixel intensity, illumination, or due to poor contrast and can't be used directly. Therefore, it is necessary to get rid of impulse noise presented inan image. In order to remove such impulse noise, Median based filters are commonly used. However, we use various types of median filters such as recursive median filter, iterative median filter, directional median filter, weighted median filter, adaptive median filter progressive switching median filter and threshold median filter. This paper will survey various median filtering techniques for excluding noisy pixel from a digital image.

Noise Suppression from images is one of the most important concerns in digital image processing. Impulsive noise may occur during image acquisition, transmission or storage Noise should be removed in such a way that important information of image should be preserved. We can use so many algorithms for getting the original image, by removing salt and pepper noise from the corrupted images. In this paper an algorithm is proposed for the restoration of gray scale images that are highly corrupted by impulse noise (salt and pepper noise). There are two phases in the proposed algorithm. First phase detects whether the processing pixel is corrupted or not. In the Second phase it recreates the corrupted pixel by means of the proposed algorithm. This algorithm shows better results than the Standard Median Filter (MF), Center Weighed Median Filter(CWM), Adaptive Median Filter(AMF), Adaptive Center Weighed Median Filter (ACWM), Decision Based Algorithm (DBA) and Modified Decision Based Unsymmetrical Trimmed Median Filter Algorithm(MDBUTMF). Obtained results with different grayscale images shows that proposed algorithm gives better Peak Signal-to-Noise Ratio (PSNR) and less Computational time and works well in removing salt and pepper noise at low, medium and high noise densities.

Impulse noise is one of the major problems in digital image processing. Impulse noise can corrupt the images and also damage the fine details in the image. Different filtering algorithms are applied on images to remove the impulse noise that are present in image during acquisition. Corrupted pixels in image take either the minimum or maximum gray level. Image denoising filters are based on the median filter. Survey on Variants of median filters like weighted median filter, adaptive switching median filter, triststate switching median filter, modified decision based unsymmentrictristate median filter are discussed in his paper.

Multi-frequency phase sensitive demodulation plays an important role in electrical impedance tomography (EIT). The phase sensitive detector is an effective tool for discriminating the amplitude and phase of each frequency component signal. It can be considered as the optimal linear filter in case of Gaussian-type noise. In real applications, however, the noise has also rather distinctive impulsive characteristics. It is a challenging task to remove the impulsive noise effectively in multi-frequency phase sensitive demodulation. In this paper, an approach for impulsive noise removal is presented in multi-frequency phase sensitive demodulation, which takes advantage of the periodicity of a multi-frequency signal and the significant properties of the median filter. The current work focuses on the use of two-dimensional (2D) median filter to handle the impulsive noise that existed in the measured signal, i.e. a 2D median filter is employed as the pre-stage signal conditioning module prior to the phase sensitive detector. The main advantages of the proposed method are that: 1) increasing the number of samples within the median filter window, which benefits the consecutive impulsive noise removal, and 2) no additional storage and computational resources are required for hardware system. Two classical methods are selected as comparisons in experiment. Numerical simulations and experimental results demonstrate that the proposed method can effectively eliminate the influence of impulsive noise on multi-frequency phase sensitive demodulation, and is able to achieve outstanding signal-to-noise ratio (SNR) performance.

We observed that certain fundamental concerns commonly exist in some state-of-the-art switching-based median filters: (i) fixed thresholding for the pre-assumed noise density, (ii) the noise decision accuracy at high density impulse noise, and (iii) the filtering scheme adopted in response to pixel characteristic type identified. In this paper, we propose a novel noise adaptive soft-switching median (NASM) filter to effectively address the above-mentioned issues and achieve much improved filtering performance in terms of efficiency in removing impulse noise and robustness against noise density variations. Experimental results also reveal that the performance of our NASM filter is fairly close to that of ideal-switching median filter.

Images are often corrupted by impulse noise from a noisy sensor or channel transmission errors. The goal of impulse noise removal is to suppress the noise while preserving the edges and the details. To this end, nonlinear techniques have been found to provide more satisfactory results in comparison to linear methods. The two often used nonlinear methods are the median filter and the outlier method. In median filtering the central pixel inside a window is replaced by the median value of all pixels in the window. However, median filtering not only removes impulse noise but also introduces distortion, in particular, for highly corrupted images. The outlier method is based on a detection-estimation strategy in which the impulse noise is detected first and then an average of all pixel values inside a window centered at the defected corrupted pixel is used as an estimate of its true value. This approach is also not effective for highly corrupted images. Recently, we proposed an efficient method to remove impulse noise from a highly corrupted image also based on a detection-estimation strategy, in which the impulse noise is detected first using a simple quadratic filter, and a selectively chosen local mean is used to estimate the true value of the corrupted pixel. A major limitation of this approach is that often it has to be applied twice to arrive at a satisfactory result. This second run not only increases computational cost but also reduce the quality of the processed image. In this paper, we analyze the limitation of this newly proposed method of impulse noise removal, and then advance a modified version which does not require a second run. >

There has been extensive research on removing universal noise from corrupted images using various kinds of non-linear filters. Median and bilateral filters are basic nonlinear filters which have been modified over the years to improve the quality of images. Median filters are good at removing Impulse noise, while these filters fail terribly in removing Gaussian noise. For Gaussian noise removal bilateral filter is preferred and successful over the years. But bilateral filter is not effective in removing impulse noise. Switching Bilateral Filter (SBF) works good for removing universal noise which involves noise detection followed by filtering. Detection plays vital role in image denoising of highly corrupted images. Boundary Discriminated Noise Detection (BDND) algorithm is very efficient in detection of noise even in highly corrupted images. Boundary Discriminated Switching Bilateral Filter (BDSBF) is proposed which employs BDND detection followed by SBF for better accuracy in noise detection even at higher noise conditions. Proposed filter also changes the size of the window based on the noise density computed from noise statistics after the BDND detection. Proposed filter shows superior results over SBF under high noise densities and high noise standard deviations.

Nonlinear vector median filters (VMF) and their variants represent ones of the most popular approaches for color image processing. In this paper a novel effective method for detection and removal impulse noise in highly corrupted color images has been proposed. First, window operator based on an explicit use of spatial relations between color image elements to detect impulse noise is used. Then the spatially connected modification of VMF for removal of previously detected impulsive noise has been employed. The performance of the proposed detector and filter for detecting and suppression of impulsive noise in test images is compared to conventional vector median algorithms.

Impulsive noise usually has adverse effect on image processing. It has been proved that adaptive median filter is effective to impulsive noise removal. However, high percentage impulsive noise is still a challenge to adaptive median filter. Thus, a sequential median filter is proposed in this paper. Adaptive median filter and median filter are combined in sequence to handle high percentage impulsive noise. Moreover, variable window sizes are employed in the sequential median filter. Experimental results indicate that the sequential median filter is effective for high percentage impulsive noise removal.

This paper presents an automatic method for simultaneous detection and elimination of impulsive noises from audio signals. The method is based on the sparse signal decomposition (SSD) with overcomplete hybrid dictionary (OHD) matrix including impulse and sinusoidal waveforms derived from discrete impulse, sine and cosine functions, and 1-norm optimization algorithm. The SSD technique decomposes an audio signal into two parts: an approximation part including the sinusoidal components; and a detail part including the impulsive and background noise components. Results show that the impulse waveforms can adequately capture clicks, pops and record scratches in audio signals meanwhile the sinusoids of the OHD matrix can effectively capture sinusoidal components. For performance comparison purpose, four restoration methods are implemented based on the representation dictionaries, median filter, and signal dependent rank order mean techniques. The methods are validated using audio signals extracted from old archive Indian and English movies and standard music database. The quality of the restored audio signals is evaluated using both subjective quality and objective quality metrics. Results show that the proposed method effectively removes impulsive noises without degrading the perceived audio quality and its intelligibility as compared to the existing restoration methods.

Impulse noise is one of the most important factors in degrading of image quality. In this paper, a novel technique is presented for detecting and removing of impulse noise, while the significant information of image, such as edges and texture, are remind untouched. The proposed algorithm use the weighted window with variable sizes and apply median filtering on them. Simulation results, with various images and noise intensities, show that the proposed algorithm has better performance compared with state of the art methods and increases the PSNR value (of the reconstructed image) up to 4dBs.

The central goal of this dissertation is to design and model a smoothing filter based on the random single and mixed noise distribution that would attenuate the effect of noise while preserving edge details. Only then could robust, integrated and resilient edge detection methods be deployed to overcome the ubiquitous presence of random noise in images. Random noise effects are modeled as those that could emanate from impulse noise, Gaussian noise and speckle noise. In the first step, evaluation of methods is performed based on an exhaustive review on the different types of denoising methods which focus on impulse noise, Gaussian noise and their related denoising filters. These include spatial filters (linear, non-linear and a combination of them), transform domain filters, neural network-based filters, numerical-based filters, fuzzy based filters, morphological filters, statistical filters, and supervised learning-based filters. In the second step, switching adaptive median and fixed weighted mean filter (SAMFWMF) which is a combination of linear and non-linear filters, is introduced in order to detect and remove impulse noise. Then, a robust edge detection method is applied which relies on an integrated process including non-maximum suppression, maximum sequence, thresholding and morphological operations. The results are obtained on MRI and natural images. In the third step, a combination of transform domain-based filter which is a combination of dual tree – complex wavelet transform (DT-CWT) and total variation, is introduced in order to detect and remove Gaussian noise as well as mixed Gaussian and Speckle noise. Then, a robust edge detection is applied in order to track the true edges. The results are obtained on medical ultrasound and natural images. In the fourth step, a smoothing filter, which is a feed-forward convolutional network (CNN) is introduced to assume a deep architecture, and supported through a specific learning algorithm, l2 loss function minimization, a regularization method, and batch normalization all integrated in order to detect and remove impulse noise as well as mixed impulse and Gaussian noise. Then, a robust edge detection is applied in order to track the true edges. The results are obtained on natural images for both specific and non-specific noise-level.