Multiple Noises Removal from Computed Tomography (CT) Images

Abstract

ABSTRACT Noise is seen in images during image acquisition and transmission. This is characterized by noise model. Image enhancement through noise reduction is a fundamental problem in image processing. Noise filtering is a necessary action and has become one of the indispensable components of image processing operation. In this work, we have applied different filtering techniques (arithmetic mean filtering, median, and Weiner) to remove multiple mixed noises such as; (speckle, Gaussian, salt and pepper) from Computed Tomography (CT) images. Performance metrics such as; Peak Signal to Noise Ratio (PSNR), and Mean Square Error (MSE) were used to evaluate filtering techniques. The results showed that the median filter had worked more effectively to remove these noises. All the above mentioned techniques were implemented using MATLAB environment. It should give some results showing the values of PSNR and MSE for these flirters to indicate the differences between them.