Noise adaptive switching median-based filter for impulse noise removal from extremely corrupted images
In this study an approach to impulse noise removal is presented. The introduced algorithm is a switching filter which identifies the noisy pixels and then corrects them by using median filter. In order to identify pixels corrupted by noise an analysis of local intensity extrema is applied. Comprehensive analysis of the algorithm performance [in terms of peak signal-to-noise ratio (PSNR) and Structural SIMilarity (SSIM) index] is presented. Results obtained on wide range of noise corruption (up to 98%) are shown and discussed. Moreover, comparison with well-established methods for impulse noise removal is provided. Presented results reveal that the proposed algorithm outperforms other approaches to impulse noise removal and its performance is close to ideal switching median filter. For high noise densities, the method correctly detects up to 100% of noisy pixels.
Available from: sciencepubco.com
- "It has an added drawback of more complex circuitry and difficulty for the determination of smoothing factor β. An efficient approach by How-Lung presents the Noise Adaptive Soft Switching Median Filter (NASSMF)  for detection and filtering. In this technique small window size is chosen for low density noise level and big window size is selected for high density noise level. "
Available from: Farokh Marvasti
- "Algorithm (DBA) , Median-based Switching filter (MS) , and Modified Decision Based Unsymmetric Trimmed Median Filter (MDBUTMF) . Also, due to the nature of impulse noise, some methods are proposed based on fuzzy logics, such as Detail-Preserving Filter (DPF) , Noise Adaptive Fuzzy Switching Median (NAFSM) filter , and Turbulent Particle swarm optimization based Fuzzy Filtering (TPFF) . "
[Show abstract] [Hide abstract]
ABSTRACT: In this paper, we propose a method for real-time high density impulse noise
suppression from images. In our method, we first apply an impulse detector to
identify the corrupted pixels and then employ an innovative weighted-average
filter to restore them. The filter takes the nearest neighboring interpolated
image as the initial image and computes the weights according to the relative
positions of the corrupted and uncorrupted pixels. Experimental results show
that the proposed method outperforms the best existing methods in both PSNR
measure and visual quality and is quite suitable for real-time applications.
Available from: Vikant Bhateja
- "This noise can corrupt images where the corrupted pixel takes either maximum or minimum gray level value; leading to severe degradation of image quality and loss of fine details -. The objective of noise suppression in such corrupted images is to filter the impulses so that the noise free image is fully restored with minimum signal distortion -. Several non-linear filters have been proposed for restoration of images contaminated by salt and pepper noise. Among them, the conventional filtering approaches include: Local Statistics & Standard Median Filter (SMF) -, Bilateral Filters ,, Decision based median filtering (DMF) approaches -, Adaptive median filters (AMF) - and Directional weighted median filters (DWMF) -. "
[Show abstract] [Hide abstract]
ABSTRACT: In this paper, a non-iterative adaptive median filter is proposed for denoising images contaminated with impulse noise. The proposed denoising scheme operates in two steps. Firstly, the pixels are segregated as 'noisy' and 'noise-free' so that the subsequent processing can be carried out only for the noisy pixels only in the next step. Secondly, the identified noisy pixels are replaced by the median value or by its neighboring pixel value. The term 'adaptive' justifies the filters' capability to increase the size of the spatial window, depending upon the decisions made based on statistical parameters (estimated within the local window). Further, the 'non-iterative' feature projects that there is no need of recursive filtering to reduce the residual noise content. The proposed denoising method is tested on images with different characteristics and is found to produce better results in terms of the qualitative and quantitative measures of the image in comparison to other filtering approaches.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.