Change Detection in SAR Images Based on the ROF Model Semi-Implicit Denoising Method

Area of Science:

• Remote Sensing
• Image Processing
• Computer Vision

Background:

• Traditional ROF models for image denoising suffer from instability and require numerous iterations.
• Synthetic Aperture Radar (SAR) image analysis requires effective denoising for accurate change detection.

Purpose of the Study:

• To introduce a novel semi-implicit denoising method for the ROF model.
• To apply this method to enhance change detection in SAR images.

Main Methods:

• The proposed ROF model semi-implicit denoising method was applied to calibrated ENVI remote sensing images.
• Difference images were generated using logarithmic ratio and mean ratio methods, followed by Principal Component Analysis (PCA) fusion.
• Fuzzy Local Information C-Means clustering (FLICM) was employed for change region identification.

Main Results:

• The semi-implicit denoising method effectively processed SAR images.
• The combined approach of ratio methods, PCA fusion, and FLICM clustering yielded accurate change detection.
• The method demonstrated high detection accuracy and operational efficiency.

Conclusions:

• The proposed ROF model semi-implicit denoising method is a robust technique for SAR image change detection.
• This method offers a significant improvement in both accuracy and computational efficiency over traditional approaches.