Smoothing and Noise Reduction in Images Using Variable Mode Decomposition

Abstract

The error between true colours of a point in a scene relates to real world, and the pixel obtained from an image sensor is known as noise. Various noises were observed, and different filtration and sub-band decomposition techniques were also achieved most success rate in reducing the noise rate. But this leads to harshness on the image pixels they were tuned to more sharp than they need. Now a novel approach discussed in next sections which helps in learning minimization of sharp noise and increases the smooth characteristics of the noisy images. The noise reduction process is carried out by an iterative decomposition method and very much dependent on intrinsic mode function (IMF) and is known as variable mode decomposition (VMD). The database comprised with medical images and synthetic aperture radar images, and the experiments were done on SAR and medical images and shown in results section. This also helps in optimising the direction method of multipliers approach.