Efficient removal of impulse noise from digital images
ABSTRACT A new impulse noise removal technique is presented to restore digital images corrupted by impulse noise. The algorithm is based on fuzzy impulse detection technique, which can remove impulse noise efficiently from highly corrupted images while preserving image details. Extensive experimental results show that the proposed technique performs significantly better than many existing state-of-the-art algorithms. Due to its low complexity, the proposed algorithm is very suitable for hardware implementation. Therefore, it can be used to remove impulse noise in many consumer electronics products such as digital cameras and digital television (DTV) for its performance and simplicity.
[Show abstract] [Hide abstract]
ABSTRACT: A novel adaptive fuzzy directional median filter is proposed in this paper which considers the directional pixels (horizontal, vertical and two diagonal directions) for estimation of adaptive threshold and incorporates the remaining background pixels based on directional statistics for efficient noise detection. The proposed filter consists of two phases: adaptive fuzzy noise detection phase followed by fuzzy filtering phase. In fuzzy noise detection phase, intensity differences from the central pixel in a 5 × 5 sliding window are calculated in four main directions, i.e., horizontal, vertical and two diagonal directions. Average value and central pixel value of 5 × 5 sliding window of newly constructed intensity differences image are exploited with fuzzy membership function to adaptively estimate threshold parameters. These parameters are then merged with fuzzy rules to detect the noise especially in detailed regions of an image. In filtering phase, simple median filter and directional median filters are smartly used based on edge and background information to restore the noisy pixels detected as noisy in adaptive fuzzy noise detection process. Experimental results based on well known quantitative measures shows the effectiveness of the proposed technique.Applied Soft Computing 04/2015; 29. DOI:10.1016/j.asoc.2015.01.010 · 2.68 Impact Factor
Conference Paper: A distributed intrusion detection model based on cloud theory[Show abstract] [Hide abstract]
ABSTRACT: Cloud computing is defined as the storage, management, processing, and accessing information and other data stored in a specific server. With the advent of internet, intrusion attacks have gained sophistication over the time. Distributed attacks could not be detected by the present available intrusion detection system. In this case, we propose a distributed intrusion detection model based on Cloud theory. Our model is composed by Intrusion Detection Agent subsystem and Data Aggregation subsystem. Intrusion Detection Agent subsystem has three parts: data collection module, Cloud decision-making module and communication module. An intrusion detection algorithm based on Cloud theory was proposed to detect intrusion behavior and improve the detection ability to complicated intrusion. Followed by our model, we introduced a strategy to defend DDoS attack using the elastic properties of cloud platform.2012 IEEE 2nd International Conference on Cloud Computing and Intelligence Systems (CCIS); 10/2012
[Show abstract] [Hide abstract]
ABSTRACT: Switching Median Filter with Boundary Discriminative Noise Detection (BDND) is one of the useful methods that are capable to restore digital images which have been extremely corrupted by universal impulse noise. Following the fundamental framework of the switching median filter, the construction of BDND can be divided into two stages. The first stage classifies the pixels into either "noise" or "noise-free" pixels, while the second stage restores the image by changing only the intensity values of the "noise" pixels. Unfortunately, the originally proposed BDND employs sorting operations in both of its stages. This condition makes the originally proposed BDND computationally expensive. Therefore, in this paper, an implementation of BDND with reduced computational time is suggested. This reduction is achieved mainly by manipulating the local histograms' properties. Experimental results show that the proposed implementation successfully produces the same results as the originally proposed BDND, but with much shorter processing time.