REMOTE SENSING
Hi, I'm Olivier STOCKER, I publish here my work concerning remote sensing. I really fell in love with this domain and all it's applications, from urban spread monitoring with sattelite images to railway 3D point cloud classification with deep learning. This site allows me to share my works in remote sensing as well as holding my paper visually enhanced version.
Discover all my internet activities via my LinkTree
News
I am currently working on the visualisation of my simulated point clouds and the associated models. If you want to see the current it is here : My cool PhD
I submitted recently to IGARSS 2023, hope to be accepted and see you all there !
Publications
Conference paper
- O. Stocker and A. Le Bris: Can SPOT-6/7 CNN semantic segmentation improve Sentinel-2 based land cover products? Sensor assessment and fusion, ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2020, 2020. ISPRS publication page. Full paper. See the enhanced visualisation page of inference results.
Communications
- O. Stocker, E. Guilbert, A Ferraz and J.F. Côté : Simulation d’acquisitions LiDAR aéroportées en milieu forestier pour l’entrainement de réseaux de Deep Learning. TriDiFor symposium, 2022. See TriDiFor 2022 programmation.
- A. Le Bris and O. Stocker : Classification par réseaux de neurones convolutifs de données SPOT-6/7 et Sentinel-2 pour la segmentation sémantique de l'occupation du sol. TEMU Workshop, 2020. See TEMU 2020 programmation.
Projects
This is the list of my MVP (Most Valuable Project):
- 2019, Master internship, ENSG & LaSTIG IGN: Use of Sentinel-2 and SPOT 6/7 data for land cover classification. Master Thesis, Oral Defense slides, Poster.
- 2019, Master project, ENSG: LiDAR semantic segmentation with random forest. Summary.
- 2017, International internship, RIEGL: Portable visualization interface for LiDAR acquisition. Summary.
- 2017, Master internship, Polytech Lille & SNCF: LiDAR post-processing for cross profiles automated generation. Summary.
Code
You can find my open source development on my GitHub:
- OS_NET: This is the full processing chain which infered the results prensented in the ISPRS 2020 paper.
- OSToolBox: A handy toolbox I developed in parallel of OS_NET containing general fonctions like file grabbers or string formaters. I released it as a Pyhton package on PyPI : OSToolBox.
- RSC: The small LiDAR classification program I developed in 1 month span during ENSG master.
Misc
Here is an old prototype of inference result visualisation I did during my intership at IGN. It is a cartography of land cover and land use at country scale "Occupation du Sol à Grande Echelle" (OCS-GE) nomenclatures over Toulouse (France) : Land cover and land use maps from SPOT VHRS
I am a 5b climber and working hard to get to finish my first 6 class route ! Just kidding, dislocated my shoulder again.
I am significant !
Contact
Feel free to reach me via stocker.olivier@gmail.com.
Original top image author: Abass; Site editor: Link to creator