The correlation technique, or template matching, applied to the detection and analysis of seismic events has demonstrated its performance and usefulness in the processing chain of the CEA/DAM National Data Center. Unfortunately, this method suffers from limitations which limit its effectiveness and its use in the operational environment, linked on the one hand to the computational cost of massive data processing, and on the other hand to the rate of false detections that could generate low-level processing. The use of denoising methods upstream of processing (example: deepDenoiser, by Zhu et al., 2020), could also increase the number of erroneous detections. The first part of the research project consists of providing a methodology aimed at improving the processing time performance of the multiplets detector, in particular by using information indexing techniques developed in collaboration with LIPADE (L-MESSI method , Botao Peng, Panagiota Fatourou, Themis Palpanas. Fast Data Series Indexing for In-Memory Data. International Journal on Very Large Data Bases (VLDBJ) 2021). The second part of the project concerns the development of an auto-encoder type “filtering” tool for false detections built using machine learning. The Syria-Turkey seismic crisis of February 2023, dominated by two earthquakes of magnitude greater than 7.0, will serve as a learning database for this study.

Thanks to satellite and airborne imagery, 3D reconstruction of earth surface is possible. Optical imagery exploits stereoscopic acquisitions and photogrammetry to retrieve 3D surface whereas interferometry is used for radar imagery. These techniques are complementary. Radar images allow the retrieval of fine metallic objects such as pylons. Optical imagery is more robust but such fine details cannot be preserved due to smoothing. An objective of the post-doctorate is to detect such fine objects.
The complementarity between 3D cloud points retrieved from satellite optical imagery and satellite and airborne radar imagery should lead to a 3D product including objects principally detected by radar and surface reconstruction derived from optical imagery.
The post-doctorate will begin with a state of the art review on 3D reconstruction by optical and radar imagery as well as cloud points fusion. Different 3D reconstruction processing chains should be used on airborne and satellite images. A precise registration algorithm and fusion algorithm on cloud points should be developed, enabling the detection of points detected only by radar. For this step, Deep Learning techniques could be useful. The results will be compared to 3D very high resolution acquired by Lidar to quantify the results quality of the proposed algorithm.
This post-doctorate will take place in labs specialized in satellite and radar image processing through a collaboration between CEA-DAM and Onera.