PolSF

Dataset Overview

PolSF is a Polarimetric Synthetic Aperture Radar (PolSAR) dataset covering the San Francisco area. It includes images from five different satellites and acquisition times.

Data Source

• Data downloaded from https://ietr-lab.univ-rennes1.fr/polsarpro-bio/san-francisco/
• Providers include CNSA, CSA, ESA, IECAS, ISRO, JAXA, MDA, NASA‑JPL, NSOAS
• Annotations performed by the IPIU team

Data Files

The dataset comprises the following files:

• SF-XXX: Contains ground‑truth label2d, PauliRGB image, and colored ground‑truth label3d
• label2dto3d.py: Script demonstrating the coloring process

Color Codes and Class Names

SF‑ALOS2

• Color codes: [[0,0,0],[132,112,255],[0,0,255],[0,255,0],[192,0,0],[0,255,255],[255,255,0]]
• Class names: 0 unlabel, 1 Mountain, 2 Water, 3 Vegetation, 4 High‑Density Urban, 5 Low‑Density Urban, 6 Developed

SF‑GF3

• Same color codes as SF‑ALOS2
• Same class names as SF‑ALOS2

SF‑RISAT

• Same color codes as SF‑ALOS2
• Same class names as SF‑ALOS2

SF‑RS2

• Color codes: [[0,0,0], [0,0,255],[0,255,0],[255,0,0],[255,255,0],[255,0,255]]
• Class names: 0 unlabel, 1 Water, 2 Vegetation, 3 High‑Density Urban, 4 Low‑Density Urban, 5 Developed

SF‑AIRSAR

• Color codes: [[0,0,0], [0,255,255],[255,255,0],[0,0,255],[255,0,0],[0,255,0]]
• Class names: 0 unlabel, 1 Mountain, 2 Water, 3 Urban, 4 Vegetation, 5 Bare soil

Reference

Xu Liu, Licheng Jiao, Fang Liu, Dan Zhang, Xu Tang. PolSF: PolSAR image datasets on San Francisco[C]//International Conference on Intelligence Science. Cham: Springer International Publishing, 2022: 214‑219.