Homogeensete objektide piiride eristamine aerofotodelt ja laserskaneerimise andmestikust näidistel põhineva masinõppe abil / Application of machine learning for the mapping of homogeneous patches from aerial images and laser scanning data
Organization
Tartu Observatoorium
Abstract
Töö eesmärgiks on põldude piiride automaattuvastuse ja -kontrollimise metoodika arendamine. Algandmeteks on Eesti Maa-ameti poolt tehtavad ortofotod ja laserskaneerimise andmed ja põllupiiride kaardikiht. Andmetöötluseks võib kasutada mistahes algoritme ja raudvara tingimusel, et pakutav lahendus oleks vajadusel rakendatav tervet Eestit katval andmestikul ja eeldatavalt kasutatakse vabavara.
The aim of the study is to develop methods for the detection of patch borders based on ortophotos and airborne laser scanning data. The solution is expected to decrease manual workload. There are no restrictions on selecting algorithms or hardware. However, the solution must be applicable for the entire country and must be based on freeware libraries.
The aim of the study is to develop methods for the detection of patch borders based on ortophotos and airborne laser scanning data. The solution is expected to decrease manual workload. There are no restrictions on selecting algorithms or hardware. However, the solution must be applicable for the entire country and must be based on freeware libraries.
Graduation Theses defence year
2018-2019
Supervisor
Mait Lang
Spoken language (s)
Estonian, English
Requirements for candidates
Level
Bachelor, Masters
Application of contact
Name
Tartu Observatoorium
Phone
E-mail
Full Document