Forest stand taxonomy using remote sensing data and artificial intelligence algorithms
The Ministry of the Environment is looking for an advanced and efficient stand segmentation technology that uses different resolutions of spectral data, polarimetric data series and stand height data from laser scanning.
Context
- The planning, organisation and control of productive activities in forests and the development of stands are carried out on the basis of forest management plots. The surveying, delimitation, digitisation and continuous updating of forest management plots in the GIS database is a crucial and human and financially demanding task, which is coordinated and carried out by public authorities and private contractors.
- All these operations, both in forests and on complex computer systems, are carried out manually, resulting in high maintenance costs for the information system and a high probability of subjective and technical errors.
Solution
- An intelligent and efficient stand segmentation technology needs to be developed, using spectral data at different resolutions, polarimetric data series and stand height data from laser scanning. The stand classification algorithm should be based on artificial intelligence technologies, testing the main methods and selecting the optimal solution. The database for the tree classification should be compatible with the boundaries of existing blocks, logging sites, forest infrastructure as well as elements of the urban, economic and natural landscapes.
Prizes for participants
Prizes will be awarded to the first, second and third place winners:
- 1st prize – 12 800 EUR
- 2nd prize – 5 000 EUR
- 3rd prize – 2 000 EUR
Additional information
The deadline for receipt of projects or applications for participation is 09.11.2021, 9:00.
For a detailed description of the challenge and the call for projects, please visit the Central Public Procurement System (CVP IS).