Perceptually Optimal Color Representation of Fully Polarimetric SAR Imagery

Area of Science:

• Remote Sensing
• Earth Observation
• Image Processing

Background:

• Polarimetric synthetic aperture radar (PolSAR) data offers rich scattering information but is perceptually challenging for direct interpretation.
• Existing SAR color representation methods lack optimal visual clarity and land cover correspondence.

Purpose of the Study:

• To develop an innovative method for generating perceptually excellent color images from fully polarimetric SAR data.
• To accurately represent scattering mechanisms and land cover types through enhanced color visualization.

Main Methods:

• A two-step coloring procedure involving Cholesky decomposition to fuse four polarimetric channels (HH, HV, VH, VV) into RGB bands.
• Transformation to the perceptually uniform CIELab color space for color equalization and distribution.
• Inverse transformation to obtain the final RGB image with enhanced visual properties.

Main Results:

• The derived color images are perceptually excellent for the human eye, resembling Google Earth maps.
• The method accurately conveys scattering mechanisms and improves color correspondence for various land cover types.
• The fused RGB images retain correlation properties, enhancing information content.

Conclusions:

• The proposed method effectively translates complex PolSAR scattering information into an intuitive and informative color image format.
• This approach offers a significant improvement over existing methods for SAR data visualization and land cover analysis.