RFI Artefacts Detection in Sentinel-1 Level-1 SLC Data Based On Image Processing Techniques.

Area of Science:

• Geoscience
• Remote Sensing
• Signal Processing

Background:

• Interferometric Synthetic Aperture Radar (InSAR) data processing is crucial for monitoring ground displacements.
• Radio-Frequency Interference (RFI) artefacts commonly contaminate InSAR data, complicating analysis and potentially leading to misinterpretations.
• The Permanent Scatterers InSAR (PSInSAR) workflow requires clean SAR image stacks for accurate results.

Purpose of the Study:

• To develop an efficient methodology for identifying and removing RFI artefacts from SAR images.
• To enhance the reliability of automatic PSInSAR processing workflows.
• To improve the accuracy of ground displacement measurements derived from InSAR data.

Main Methods:

• Implementation of image processing techniques for RFI detection.
• Utilization of feature extraction methods including pixel convolution, thresholding, and nearest neighbor structure filtering.
• Application of a convolutional neural network as a reference classifier for RFI identification.

Main Results:

• Successful development of a methodology to mark and remove images contaminated by RFI artefacts.
• Demonstrated efficiency of image processing techniques in detecting RFI.
• Provided a foundation for automated RFI artefact removal in PSInSAR workflows.

Conclusions:

• The proposed RFI detection and removal methodology effectively addresses data contamination in InSAR processing.
• This technique supports the automation of PSInSAR workflows by ensuring data quality.
• Accurate ground displacement monitoring is enhanced through the mitigation of RFI artefacts.