Artistic Edge and Corner Enhancing Smoothing Giuseppe Papari Nicolai Petkov Patrizio Campisi.
-
date post
21-Dec-2015 -
Category
Documents
-
view
218 -
download
2
Transcript of Artistic Edge and Corner Enhancing Smoothing Giuseppe Papari Nicolai Petkov Patrizio Campisi.
ABSTRACT 1) absence of texture details
2) increased sharpness of edges as compared to photographic images
generalizes both the well known Kuwahara filter and the more general class of filters known as VCFS.
VCFS : value and criterion filter structureValue-and-criterion filters have a `value' function (V) and a `criterion' Value-and-criterion filters have a `value' function (V) and a `criterion' function (C), each operating separately on the original image, and a function (C), each operating separately on the original image, and a `selection' operator (S) acting on the output of C. The selection `selection' operator (S) acting on the output of C. The selection operator chooses a location from the output of C, and the output of V operator chooses a location from the output of C, and the output of V at that point is the output of the overall filter.at that point is the output of the overall filter.
INTRODUCTION Linear low-pass filtering strongly
attenuates high-frequency components, not only noise, but also edges and corners, are smoothed out.
There has been a remarkable effort to find a nonlinear operator able to remove texture and noise, while preserving edges and corners.
ECPS : edge and corner preserving smoother
Ex : median filtering, morphological analysis, bilateral filtering
INTRODUCTION
current work : In a specific aspect of ECPSs, their ability to
produce images that are visually similar to paintings. algorithm makes use of :
1) a different set of weighting subregions
for computing local averages 2) a different combination criterion which generalizes the minimum standard deviation rule and which does not suffer the above mentioned ill-posedness.
KUWAHARA FILTER AND EXTENSIONS
A. Review of the Kuwahara Filter
MSDC :
minimum standard deviation criterion
EXPERIMENTAL RESULTS
A. Comparison With Existing Approaches Fig.8Fig.9Fig.10Fig.11Fig.12Fig.13Fig.14