DATA ANALYSIS

Usage of adaptive homomorphic filters for CT processing

M. S. Usanov, N. S. Kulberg, S. P. Morozov

SPC of Medical Radiology of the Moscow Department of Healthcare

Abstract: The article deals with effective usage of homomorphic filtering in the processing of data that do not comply with Gaussian distribution law. As an example, we used data from computed tomography. Data processing is based on wavelet – filtering, including noise reduction and edge enhancement for objects of interest. Results have shown significant data quality enhancement. What’s more important, we managed to avoid some unacceptable artifacts that occur when processed without the suggested transformation.

Keywords: computed tomography, image processing, nonlinear transform, edge enhancement, noise reduction inverse transform sampling, homomorphic filters.