Psst, remember the cheatsheet!

Research usecases for Puhti

Contents

Research usecases for Puhti#

Check out what others have achieved using CSC supercomputers:

Some interesting usecases from CSC seminars#

Path from drone imagery to they can be processed using Puhti batch job with OpenDronemap — Jon Atherton, UH: Optical sensing and 4D modelling of plant ecophysiological traits#

Two point clouds of trees with their extracted stem point cloud — Jiri Pyörälä, UH, NLS/FGI: Experiences with PCLpy on Puhti#

Example result of land cover classification in Northern Finland, computed with sklearn and dask — Mikko Impiö, SYKE: Modern machine learning for land cover classification#

Sentinel-2 true color mosaic with its single components in time and space — Arttu Kivimäki, FGI/NLS: Mosaicking Sentinel-2 data in Puhti#

Coarse water segmentation probability map based on 512x512 pixel batches, which is too coarse, but too much context would require more computational resources — Tapio Friberg, ICEYE: LUMI usecase#

You can find more use case presentations from CSC: geocomputing seminars page.

Some publications from Finland that used Puhti#

Crop yield predicton workflow from image download to prediction — Maria Yli-Heikkilä et al, LUKE: Scalable Crop Yield Prediction with Sentinel-2 Time Series and Temporal Convolutional Network#

Example results from OpenDroneMap image processing — Gbagir et al, UEF: OpenDroneMap: Multi-Platform Performance Analysis#

Sentinel-1 change detection results — Eetu Jutila Masters thesis, Aalto: Land cover change detection using Sentinel-1 satellite images#

Workflow diagram for deep learning on point clouds — Andras Balazs et al, LUKE:Comparison of neural networks and k-nearest neighbors methods in forest stand variable estimation using airborne laser data#

Workflow diagram of EODIE — Samantha Wittke et al, FGI/Aalto EODIE - Earth Observation Data Information Extractor#

Know more? -> Please let us know :)