ARTICLE IN PRESS JID: CAEE [m3Gsc;January 17, 2018;15:3] Computers and Electrical Engineering 000 (2018) 1–10 Contents lists available at ScienceDirect Computers and Electrical Engineering journal homepage: www.elsevier.com/locate/compeleceng Different applied median filter in salt and pepper noise  U ˘ gur Erkan a , Levent Gökrem b,∗ , Serdar Engino ˘ glu c a Department of Computer Programing, Erbaa Vocational School, Gaziosmanpa ¸ sa University, Erbaa, Tokat 60500, Turkey b Department of Mechatronics Engineering, Faculty of Engineering and Natural Sciences, Gaziosmanpa ¸ sa University, Tokat 60100, Turkey c Department of Mathematics, Faculty of Arts and Sciences, Çanakkale Onsekiz Mart University, Çanakkale 17100, Turkey a r t i c l e i n f o Article history: Received 3 July 2017 Revised 15 January 2018 Accepted 15 January 2018 Available online xxx Keywords: Image denoising Impulse noise Salt and pepper noise Noise removal Median filter a b s t r a c t In this paper, we proposed a new method, Different Applied Median Filter (DAMF), to re- move salt and pepper (SAP) noise at all densities. We then explained some basic notions of it. Afterwards, we compared the results of DAMF method and some other methods by using Peak Signal to Noise Ratio (PSNR) and Structural Similarity (SSIM) for some images such as Cameraman and Lena. For example, for Cameraman image with a SAP noise ratio of 30%, PSNR and SSIM results of PSMF, DBA, MDBUTMF and NAFSM methods are 28.27/ 29.28/ 29.44/ 32.09 and 0.9044/ 0.9324/ 0.7740/ 0.9494 respectively while PSNR and SSIM results of DAMF method are 36.83 and 0.9844, respectively. We finally showed that DAMF could be successfully removed SAP noise at all densities. © 2018 Elsevier Ltd. All rights reserved. 1. Introduction In image processing, much research has been done on image denoising in recent years [1]. In the field of image process- ing, one of the most important topics is to remove the noise from the images. Image denoising aims to obtain the nearest image to the real one by removing the noise [2]. Image denoising, a pre-process in image processing, protects edges, tex- tures and other details which are among its most essential tasks [3]. In other words, the success of image denoising affects the success rate of the segmentation, classification and similar procedures [4,5]. As long as images are entirely not produced on computers, image noise will be a by-product of the image capturing thanks to the sensors [6]. Images are distorted by impulse noise for different reasons such as malfunctioning pixels in camera sensors or faulty memory locations in hardware. The impulse noise has two known types, which are salt and pepper noise (SAP) and random valued noise. When the image is distorted by SAP noise, corrupted pixels take the maximum and minimum value [7–10]. The most common filters for removing SAP noise are nonlinear [11]. The most common one of these nonlinear filters is the Median Filter (MF) and its derivatives. MF is effective at low-density noise [12,13]. MF’s disadvantage is that it processes all the pixels rather than only noisy ones [14] and that it uses fixed window size. However, probably the most significant downside of it is that it removes thin lines or edges and blurring image detail even at a low-density noise [15]. The filters with adaptive window size such as Adaptive Median Filter (AMF) [16] have been developed to deal with this problem. These filters have achieved lots of success compared to the others in removing high-density SAP noise. However, the problem here is that if the window size is chosen too large, it denies us to find the original pixel information [17].  Reviews processed and recommended for publication to the Editor-in-Chief by Area Editor Dr. E. Cabal-Yepez. ∗ Corresponding author. E-mail address: levent.gokrem@gop.edu.tr (L. Gökrem). https://doi.org/10.1016/j.compeleceng.2018.01.019 0045-7906/© 2018 Elsevier Ltd. All rights reserved. Please cite this article as: U. Erkan et al., Different applied median filter in salt and pepper noise, Computers and Electrical Engineering (2018), https://doi.org/10.1016/j.compeleceng.2018.01.019