In this study the experimental comparison is made with four different methods for the removal of noise that can occur in Cone Beam Computed Tomography (CBCT) images. Image information of 300 different sections obtained from 30 different patients who came to the Karadeniz Technical University Faculty of Dentistry Department of Oral Diagnosis and Radiology Clinic for routine controls became the dataset of our work. Whether there was noise information previously in selected sections that were chosen for use in the study have been measured using noise level estimation method. Different levels of noise have been added to those noise-free images and by using anisotropic diffusion, non-local means, Wiener filter and block matching 3D filter methods those noise information were tried to be cleaned. Best performing noise removal method on CBCT images have been determined comparing peak signal to noise ratio values derived from results.