A hybrid of tropical-singular value decomposition method for salt and pepper noise removal

Authors: ACHMAD ABDURRAZZAQ, ISMAIL MOHD, AHMAD KADRI JUNOH, ZAINAB YAHYA

Abstract: The unknown information contained in an image that causes the change of information in the image is called noise. In this paper, we propose a new method for removing salt and pepper noise by using singular value decomposition and the concept of tropical algebra operations. To determine the performance of the proposed method, 20 test images are used as samples. Then three different image quality assessments are used: peak signal-to-noise ratio (PSNR), structural similarity (SSIM), and image enhancement factor (IEF). In addition, six different filtering methods, i.e. MF, DWMF, PSMF, MDBUTM, NAFSM, and BPDF, are used to compare the performance of the proposed method. The experimental results show that the proposed method yields better results than the existing methods, in particular the BPDF filtering method.

Keywords: Noise removal, salt and pepper noise, impulse noise, singular value decomposition, tropical algebra

Full Text: PDF