A finite mixtures algorithm for finding proportions in SAR images.

Area of Science:

• Remote Sensing
• Image Analysis
• Signal Processing

Background:

• Synthetic Aperture Radar (SAR) images are crucial for environmental monitoring.
• Estimating class proportions in SAR imagery is essential for accurate analysis.
• Existing methods may not directly provide class proportions.

Purpose of the Study:

• To develop and present an algorithm for estimating class proportions in SAR images.
• To utilize a finite mixtures approach for direct proportion estimation.
• To apply the algorithm to aircraft SAR images of sea ice.

Main Methods:

• Assuming images consist of a mixture of known pixel types.
• Employing the Expectation-Maximization (EM) algorithm for maximum likelihood estimation.
• Applying a finite mixtures model to SAR image data.

Main Results:

• Direct estimation of class proportions from SAR images.
• Successful application to aircraft SAR images of sea ice.
• Demonstration of tailored image displays based on mixture components.

Conclusions:

• The finite mixtures approach offers a direct method for estimating class proportions in SAR images.
• The algorithm provides valuable insights for sea ice analysis.
• The method enables customized image visualizations for users.