Automated geospatial feature generation via deep learning

Automated geospatial feature generation via deep learning methods

Identyfikator grantu: PT01195

Kierownik grantu: Samuel Sisay Tadesse

Realizatorzy:

  • Samuel Sisay Tadesse

Politechnika Gdańska

Wydział Elektroniki, Telekomunikacji i Informatyki

Gdańsk

Data otwarcia: 2024-10-16

Planowana data zakończenia grantu: 2027-10-16

Streszczenie grantu

Currently conducting a research for PhD thesis, which focuses on the vectorization of satellite images. This research aims to convert complex satellite imagery into structured vector data, which can be utilized for various applications, including urban planning, agriculture monitoring, and disaster management. To achieve this, we need to train machine learning models on a substantial amount of satellite image data. Access to GPU resources is crucial for efficiently processing these large datasets and significantly speeding up the training process required for the research.

Publikacje

  1. Samuel Tadesse & Marcin Kulawiak, assessing the cross-region knowledge transfer capability of selected deep learning building vectorization methods in the context of available training datasets, Scientific Reports 60735, (2026) 22
  2. Samuel Tadesse & Marcin Kulawiak, Remote sensing imagery domain adaptation by chromatic data transfer for building vectorization, ISD 2026 — the 34th International Conference on Information Systems Development N/A, (2026) 12

Kontakt

ul Traugutta 75, 80-221 Gdańsk
tel.: + 48 58 347 24 11
email: office@task.gda.pl
NIP: 584-020-35-93
REGON: 000001620
Godziny otwarcia: pn-pt godz. 8:00-15:00