Empirical mode decomposition and adaptive bilateral filter approach for impulse noise removal

In this article, we introduced a novel algorithm to identify and correct images affected by impulse noise. The proposed technique composed of two stages: noisy pixels identification and restoration of them. Here, empirical mode decomposition is used to identify the pixels affected by impulse noise and adaptive bilateral filter is used to restore those noisy pixels. Mean absolute difference of the intrinsic mode functions (IMFs) are compared with the two dimensional cross entropic threshold value in order to identify the pixels affected by the impulse noise. In the next stage of the processing, an adaptive bilateral filter is used to retain the fine details and remove the noisy components in the image. The performance of the proposed scheme is evaluated on different benchmark images. Four performance evaluation measures: Peak signal to noise ratio (PSNR), Image Enhancement Factor (IEF), Mean Structure Similarity Index (MSSIM) and Correlation Factor (CF) are used to test the performance of the proposed algorithm. The simulation results of the proposed algorithm claim better result than the other existing state-of-the-art algorithms. (C) 2018 Elsevier Ltd. All rights reserved.