HIIT funded Research Harnesses Drones for Forestry and Agriculture
Drones can revolutionize the forestry and agriculture industries through many applications, including health monitoring, pest detection, inventory, and irrigation management. Essentially, drones serve as platforms that carry sensors. It is these sensors, such as multispectral cameras, that enable the detection and potential remediation efforts necessary to advance forestry and agriculture. For any application to be successful, it is crucial that sensor data is both robust and reliable.
Machine learning and data science can significantly enhance the quality of drone sensor data. This was the focus of Jani Kuurasuo’s Master’s thesis, which was jointly supervised by Dr. Jon Atherton from the University of Helsinki’s Department of Forest Sciences and Dr. Martha Arbayani Zaidan from the Department of Computer Science.
Over the last 12 months, Jani has worked extensively with drone imagery of forests and crops. He explains that:
“I have spent the last year or so getting to grips with drone data collection and processing pipelines to produce reflectance estimates from multispectral drone data. The main challenge and reward for me was learning about a completely new field. The practical applications such as phenotyping are also a big bonus working in this field”.
Now that his data processing work is done, Jani aims to finish writing his thesis during Autumn. The reward for his hard work is a fully funded position in the new AI Doctoral Pilot which he starts in January 2025 on a similar topic, combining AI with drone and remote sensing data.
Contact Person:
Dr Jon Atherton
Academy Research Fellow
Optics of Photosynthesis Lab,
Forest Sciences/INAR
University of Helsinki
homepage: https://www.mv.helsinki.fi/home/joathert/
dronepage: https://www.mv.helsinki.fi/home/joathert/drones/
Read more news
How to ensure anonymity of AI systems?
When training artificial intelligence systems, developers need to use privacy-enhancing technologies to ensure that the subjects of the training data are not exposed, new study suggests.
Arto Hellas receives the Nokia Foundation teaching recognition award
Arto Hellas was awarded the inaugural Nokia-OKKA Educational Recognition Award for his long-term efforts in advancing ICT education.
GRADSTOP: Early Stopping of Gradient Descent via Posterior Sampling presented at ECAI 2025
HIIT Postdoc Katsiaryna Haitsiukewich presented a full paper at ECAI-2025