Supervisor : Florence Tupin
Nombre d’étudiants : 1
SAR images are acquired by aerial or satellite sensors in all-weather and all-time conditions. Being a coherent imaging system, images are highly noisy and difficult to segment.
The objective of this project is to exploit a patch-based denoising method called NL-SAR to define a segmentation approach. Indeed, this algorithm provides an equivalent number of looks map that could be exploited to define edges and build a segmentation approach.
Programing : Matlab ou C/C++
Links : Paper and algorithm NL-SAR on the web page:
http://www.math.u-bordeaux1.fr/~cdeledal/nlsar.php
http://www.math.u-bordeaux1.fr/~cdeledal/nlsar.php