FOSS4G 2023

Lorik Haxhiu

Lorik Haxhiu, PhD. has over two decades of experience in Kosovo’s energy and natural resource sectors, where he has taken on various roles and successfully carried out technical assistance programs and projects funded by USAID, EU, World Bank and other international organizations.

GIS integration is vital for his projects as it supports them in every stage from development to planning to implementation and delivery, offering valuable outcomes for beneficiaries and stakeholders.

He currently runs a small consulting firm that specializes in environmental remediation and sustainable energy solutions.

The speaker's profile picture

Sessions

06-30
16:00
30min
Revolutionizing Solar Potential Assessments in Kosovo Using Drones and Machine Learning
Lorik Haxhiu

Imagine a future where entire communities can harness the power of the sun to fuel their homes and businesses, reducing their dependence on traditional energy sources and helping to build a more sustainable world. At FOSS4G, I am excited to share with you a groundbreaking project that is making this vision a reality in Kosovo, using the latest geospatial technology.

Through the USAID funded Kosovo Energy Security of Supply (KESS) activity, DT global is working to promote sustainable energy solutions in Kosovo. A partnership between DT Global and DevGlobal, are leveraging the power of drones, GIS software, and open-source machine learning models to revolutionize the way we evaluate the solar potential of individual structures. By accurately delineating the boundaries of rooftops using drone imagery, we can then apply cutting-edge photogrammetry analytics to determine the optimal placement of solar panels.

But we're not stopping there. By training the Ramp* open buildings model to successfully identify and delineate rooftops in Kosovo, using data obtained from the Kosovo Cadastral Agency's 2023 high-resolution aerial survey campaign, we are laying the groundwork for a national-level approach to mapping building footprints that can be utilized for a range of applications beyond evaluating rooftop solar potential.

*Ramp is an open-source machine learning model and toolset for extracting building footprints from high-resolution satellite imagery at scale.

Outdoor Stage