mosfet69 / polsar-image-denoising

A super fast denoising method for polarimetric synthetic aperture radar (POLSAR) images.

POLSAR-image-denoising

This repo contains the C/Matlab codes for the super-fast noise filtering algorithm for polarimetric synthetic aperture radar (POLSAR) images as proposed in the following paper:

Yi Cui, Yoshio Yamaguchi, Hirokazu Kobayashi, and Jian Yang. "Filtering of polarimetric synthetic aperture radar images: A sequential approach." In Geoscience and Remote Sensing Symposium (IGARSS), 2012 IEEE International, pp. 3138-3141. IEEE, 2012.

A demo to use the code to filter synthetic POLSAR images can be found in demo.m. The core functions fPOLSARSMAP.c and fPOLSARENL.c are written in C in order to leverage the speed advantage. However, for most users of the POLSAR community, the Matlab function polsar_filter.m should be the most straightforward API to apply the filter to cleaning their own data.

The following figure shows an examplary result of our filter when applied to a real POLSAR image. On the left is the noisy image whereas on the right is the filtered image. The size of the image is 2500×2500. Filtering such a large image only takes 0.87 seconds on a mediocre PC. Super fast!