S1DENOISE-1.4.0

S1DENOISE-1.4.0 can process historical Sentinel-1 data that come without ESA denosing vectors.
Changelog:

• Implement NERSC_TG algorithm: noise coefficient are computed using total gain computed from calibration and annotation metadata
• Replace XML ETree with BautifulSoup
• Drop Nansat

S1DENOISE-1.3.3

• Mask away DN=0 values (borders of images)
• Adap parameters in remove_texture_noise to processing of AI4Arctic data

S1DENOISE-1.3.2

• Use zipfile instead of unzip
• Add landmask functionality to s1_correction.py

S1DENOISE-1.3.1

Using https://sar-mpc.eu for downloading AUX_CAL data

S1DENOISE-1.2.5

Contributions from users:

• New URL for calibration files download
• Compatibility with Windows

S1DENOISE-1.2

• A new class for fast noise removal is created

S1DENOISE-1.1

Changelog:

• GPLv3
• Safe opening of XML files
• Noise scaling and power balancing coefficients for IPF 3.1 for S1A/S1B, EW/IW, HV/VH
• Coefficients are stored in JSON files
• New texturalNoiseRemoval2
• Estimate range quality metric
• Estimate azimuth quality metric
• Scripts for plotting validation summary and tables
• New design of training scripts (AnalyzeExperiment class in utils)
• New script s1_correction.py for thermal and texture and angular correction
• Improved documentation for training and denoising
• List of training files

S1DENOISE-1.0

Initial version of S1DENOISE code developed J.-W. Park and co-authors