Mohammed-Yassine Habibi

Mohammed-Yassine Habibi

PhD student in interpretable machine learning & concept learning for ecology · Inria (Thoth) & LECA · Corps des IPEF

Green’s Function Estimation from Ambient Noise with Limited Illumination in 3D

Winter 2025 · MVA — Inverse Problems and Imaging (Prof. Josselin Garnier)

Code

  • Inverse problems
  • Imaging

Project of the MVA course Inverse problems and imaging: stochastic and statistical approaches taught by Prof. Josselin Garnier.

In passive imaging, the cross-correlation of ambient noise signals recorded at two sensors can be used to estimate the Green’s function between them — as if one sensor were an active source. This project studied this estimation in 3D when the noise sources only partially illuminate the medium, and how the source configuration affects the reconstruction. See the project statement and the lecture notes.

Three 3D sensor and noise-source configurations
The three source/sensor configurations studied in 3D.
Cross-correlations of noise signals
Cross-correlations of the recorded noise signals for the three configurations.