Conference Paper (published)

Comparison of target detectors to identify icebergs in Quad- Polarimetric SAR Alos-2 images

Details

Citation

Bailey J, Marino A & Akbari V (2021) Comparison of target detectors to identify icebergs in Quad- Polarimetric SAR Alos-2 images. In: International Geoscience and Remote Sensing Symposium (IGARSS). 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, Brussels, Belgium, 11.07.2021-16.07.2021. Piscataway, NJ, USA: Institute of Electrical and Electronics Engineers Inc. pp. 5223-5226. https://doi.org/10.1109/IGARSS47720.2021.9554230

Abstract
Icebergs represent hazards to ships and maritime activities and therefore their detection is essential. Synthetic Aperture Radar (SAR) satellites are very useful for this, due to their capability to acquire under cloud cover and during polar nights. Additionally, polarimetry has been proven to improve the detection capability. In this work, we compare six state-of-the-art quad polarimetric detectors to test their performance and ability to detect small sized icebergs in four locations in Greenland. These were the polarimetric notch filter (PNF), polarimetric match filter (PMF), polarimetric whitening filter (PWF), optimal polarimetric detector (OPD), reflection symmetry detector, and the dual polarisation anomaly detector (iDPolRAD). We use four single look complex ALOS-2 quad pol images. The data were calibrated and processed. We produce the covariance matrices of each image before applying a testing and training window for detection. We also add a guard window to reduce false alarms. Our results show that the multi-look polarimetric whitening filter and optimal polarimetric detector provide the most optimal performance in quad and dual pol mode detection.

Keywords
Training; Matched filters; Satellites; Oceans; Detectors; Reflection; Marine vehicles

StatusPublished
Publication date31/12/2021
Publication date online12/10/2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Place of publicationPiscataway, NJ, USA
ISSN of series2153-7003
eISBN978-1-6654-0369-6
Conference2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS
Conference locationBrussels, Belgium
Dates

People (2)

Dr Vahid Akbari

Dr Vahid Akbari

Lect in Artificial Intelligence/Data Sci, Computing Science and Mathematics - Division

Dr Armando Marino

Dr Armando Marino

Associate Professor, Biological and Environmental Sciences