Share Email Print

Optical Engineering

Fractional order mean in image processing
Author(s): Patricia Laura Torroba; Nelly Lucia Cap; Hector Jorge Rabal; Walter D. Furlan
Format Member Price Non-Member Price
PDF $20.00 $25.00

Paper Abstract

We explore the features of a family of nonlinear filters called the fractional order mean. The median, the mode, and other statistical characteristics belong to this family as particular integer orders. The behavior of these filters in noise suppression in images is investigated. An improved version of these filters, which we call the 8-pixel window mean filter is also explored. In this case, the filtered value of the pixel is compared with its experimental value in the presence of a preset threshold. Noise suppression performance is tested with the usual mean square error (MSE) criterion, with a correlation criterion, and with a new criterion, the mean relative error (MRE). Evidence is found that the MRE correlates better than the MSE with the perceived visual quality of the processed image.

Paper Details

Date Published: 1 February 1994
PDF: 7 pages
Opt. Eng. 33(2) doi: 10.1117/12.152006
Published in: Optical Engineering Volume 33, Issue 2
Show Author Affiliations
Patricia Laura Torroba, Ctr. De Investigaciones Opticas (CIOp) (Argentina)
Nelly Lucia Cap, Univ. of La Plata (Argentina)
Hector Jorge Rabal, Centro de Investigaciones Opticas (Argentina)
Walter D. Furlan, Ctr. de Investigaciones Opticas (CIOp) (Spain)

© SPIE. Terms of Use
Back to Top