Multi-SAR: Segmentation and Classification of Multi-temporal, Multi-sensor Synthetic Aperture Radar (SAR) Data

• Research unit: INNOSUISSE
• Project number: 8466.1;6 ESPP-ES

Short description (English)

Multi-SAR: Segmentation and Classification of Multi-temporal, Multi-sensor Synthetic Aperture Radar (SAR) Data

Short description (French)

Multi-SAR: Segmentation and Classification of Multi-temporal, Multi-sensor Synthetic Aperture Radar (SAR) Data

Abstract (English)

In order to capture the information content provided by spaceborne SAR systems, sophisticated algorithms must be developed. The purpose of this project is to develop, based on SARscape® - a commercial software integrated to Geographic Information Systems (Arcview® and Envi® ) - dedicated algorithms to analyse multi-temporal, multi-sensor SAR data for the detection and extraction of land cover and land cover changes. The proposed algorithms will include tailored made speckle filtering, edge detection, region segmentation, and time varying segmentation algorithms, allowing to exploit the multi-temporal information provided by different operational spaceborne SAR data.

Abstract (French)

In order to capture the information content provided by spaceborne SAR systems, sophisticated algorithms must be developed. The purpose of this project is to develop, based on SARscape® - a commercial software integrated to Geographic Information Systems (Arcview® and Envi® ) - dedicated algorithms to analyse multi-temporal, multi-sensor SAR data for the detection and extraction of land cover and land cover changes. The proposed algorithms will include tailored made speckle filtering, edge detection, region segmentation, and time varying segmentation algorithms, allowing to exploit the multi-temporal information provided by different operational spaceborne SAR data.