سال انتشار: ۱۳۹۰
محل انتشار: هفتمین کنفرانس ماشین بینایی و پردازش تصویر
تعداد صفحات: ۵
Seyyed Reza Sharafinezhad – Department of ECEShahid Beheshti UniversityTehran, Iran
Habib Alizadeh – Department of ECETarbiat Modares UniversityTehran, Iran
Mohammad Eshghi – Department of ECEShahid Beheshti UniversityTehran, Iran
In this paper, a new method is proposed to remove thenoise from images, which have problems as mode mixing, lack oflocal maximum and minimum, blurred edges and details. Thismethod is based on adaptive ensemble empirical modedecomposition two-dimensional (AEEMD2D).First, the original image that infects using noise and interferencesignals is summed to a sequence of two-dimensional Gaussiannoise. Then, images adding to noise are decomposed into theirIntrinsic Mode Function (IMF)-s and Residue using traditionalBEMD. Amplitude of white Gaussian noise added to image and theensemble number of noises are obtained using an adaptive methodwhich is presented in Section 3. This number of ensemble noises isminimum value in comparison to other algorithm. Final IMFs andimage without noise are achieved by averaging of pervious IMFs. Acomputer simulation is used to evaluate the performance of theproposed method in comparisons to the performances otheralgorithm, as traditional BEMD.