An Algorithm for Segmentation and Detection of Changes in Signals Using Diffusion Equations

Aline R. Gesualdi, Marcelo P. de Albuquerque, Márcio P. de Albuquerque, Marcello L. R. de Campos

DOI: 10.14209/its.2002.490

Evento: 2002 International Telecommunications Symposium (ITS2002)

Keywords:

Abstract

"We will present, in details, an algorithm for the Stabilized Inverse Diffusion Equation (SIDE) technique introduced in [4]. The algorithm will be evaluated in two different application areas of interest, which are: signal detection and image segmentation. SIDEs are a family of semi-discrete evolution equations which stably sharpen edges and suppress noise. This characteristic makes SIDEs an efficient tool in noise reduction problems. This paper will focus on the detection of 1-D signals and on the segmentation of ultrasound and mammography images. Results will be presented."

Download