%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2019/09.25.22.43
%2 sid.inpe.br/sibgrapi/2019/09.25.22.43.25
%@doi 10.1109/SIBGRAPI.2019.00016
%T Low-Dose CT Dental Image Denoising by Morphological Operators and 3D Filtering
%D 2019
%A Stringhini, Romulo Marconato,
%A Welfer, Daniel,
%A d'Ornellas, Marcos Cordeiro,
%A Gamarra, Daniel Fernando Tello,
%@affiliation Federal University of Santa Maria, Brazil
%@affiliation Federal University of Santa Maria, Brazil
%@affiliation Federal University of Santa Maria, Brazil
%@affiliation Federal University of Santa Maria, Brazil
%E Oliveira, Luciano Rebouças de,
%E Sarder, Pinaki,
%E Lage, Marcos,
%E Sadlo, Filip,
%B Conference on Graphics, Patterns and Images, 32 (SIBGRAPI)
%C Rio de Janeiro, RJ, Brazil
%8 28-31 Oct. 2019
%I IEEE Computer Society
%J Los Alamitos
%S Proceedings
%K Low-dose, Computed tomography, Noise reduction, Mathematical Morphology, BM3D, PSNR.
%X The impact in reducing the radiation dose in computed tomography (CT) exams is directly related to the quality of the images obtained in these exams. Such images are degraded by undesirable artifacts, known as noise. In order to improve the quality of these images and provide an accurate medical diagnosis, it is necessary to apply noise reduction techniques. In this study, a method based on structural segmentation and filtering through morphological operators along with a BM3D filtering is proposed to reduce noise and preserve details in low-dose CT dental images. Experimental results of the proposed method were compared with several existing methods and validated using the PSNR, SSIM, MSE and EPI metrics. Our method demonstrated superior performance among the evaluated filters. In comparison to the filter that obtained the best results, our method had a gain of 12.46% on PSNR, 11.11% on SSIM, 14.5% on MSE and 9.63% on EPI metrics.
%@language en
%3 Paper ID 12.pdf