Skip to content

SurajK7/PS-pixels-detection-PSInSAR

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 

Repository files navigation

PS pixels detection PSInSAR

Neural Network-based approach for identification of PS pixels in PSInSAR.

Persistent Scatterer Interferometric Synthetic Aperture Radar (PSInSAR) is a remote sensing method, which uses isolated radar scatterers that maintain high coherence for long periods such as boulders, tree trunks, and buildings.

To identify such pixels from contaminated interferograms, we investigate a variety of deep learning approaches:

  • CNN-based approach to capture the spatial nature of noise in interferograms and directly map the contaminated phases to its Phase consistency.
  • ANN-based approach considering only the temporal variations in amplitude and phase angles of pixels.
  • CNN-based auto-encoders to denoise the interferograms, followed by an LSTM based architecture to map these cleaned interferograms to their Phase consistencies.

References:

  • S. Mukherjee, A. Zimmer, N. K. Kottayil, X. Sun, P. Ghuman and I. Cheng, "CNN-Based InSAR Denoising and Coherence Metric," 2018 IEEE SENSORS, New Delhi, 2018, pp. 1-4.
  • The simulated images dataset was generated by simulator.

About

CNN-LSTM based PS pixels detection

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published