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.

OUTLINEOUTLINE

INTRODUCTION KUWAHARA FILTER AND EXTENSIONS PROPOSED OPERATOR EXPERIMENTAL RESULTS

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

KUWAHARA FILTER AND EXTENSIONS

A. Review of the Kuwahara Filter

MSDC :

minimum standard deviation criterion

KUWAHARA FILTER AND EXTENSIONS

B. Limitations of the Kuwahara Filter Fig.2

PROPOSED OPERATOR

PROPOSED OPERATOR

MSDC :

minimum standard deviation criterion

PROPOSED OPERATOR

EXPERIMENTAL RESULTS

A. Comparison With Existing Approaches Fig.8Fig.9Fig.10Fig.11Fig.12Fig.13Fig.14

EXPERIMENTAL RESULTS

B. Influence of the Parameters

EXPERIMENTAL RESULTS

B. Influence of the Parameters

EXPERIMENTAL RESULTS

B. Influence of the Parameters

Thank you for your listening!

2007.12.18