The main purpose of this thesis is to identify macroplastics along a stretch of the Brindisi coast using photogrammetric images collected from Remotely Piloted Aircraft Systems (RPAS). After being acquired, the photos were imported into the Metashape environment and processed using Structure from Motion and Multi-View Stereo algorithms. This allowed the generation of reliable digital elevation models and accurate orthophotos concerning the study area. This information was subsequently used as input for mathematical models introduced to expediently and effectively identify macroplastics, improving the results obtained through visual on-site surveys. From this analysis, it was possible to deduce, following the manual detection of macroplastics in QGIS, the surface covered by plastic waste and the abundance index; these results were then used to generate thematic maps.
Lo scopo principale di questa tesi è di individuare le macroplastiche presenti lungo un tratto della costa brindisina mediante l’elaborazione di immagini fotogrammetriche collezionate da Remotely Piloted Aircraft Systems (RPAS). Dopo essere stati acquisiti, i fotogrammi sono stati importati in ambiente Metashape ed elaborati sfruttando gli algoritmi Structure from Motion e Multi View Stereo. Ciò ha permesso di generare modelli digitali delle elevazioni affidabili ed ortofoto accurate dell’area di studio. Tali informazioni sono state successivamente utilizzate come input dei modelli matematici introdotti per individuare le macroplastiche in modo speditivo ed efficace migliorando i risultati ottenuti mediante rilievo visivo in situ. Da questa analisi è stato possibile ricavare, dopo il riconoscimento manuale delle macroplastiche in QGIS, la superficie ricoperta da rifiuti e l’indice di abbondanza; successivamente sono stati utilizzati questi risultati per generare carte tematiche.
STRUCTURE FROM MOTION E MULTI VIEW STEREO PER L'INDIVIDUAZIONE DI MACROPLASTICHE DA RPAS NELLE AREE COSTIERE
LONERO, MARCO
2022/2023
Abstract
The main purpose of this thesis is to identify macroplastics along a stretch of the Brindisi coast using photogrammetric images collected from Remotely Piloted Aircraft Systems (RPAS). After being acquired, the photos were imported into the Metashape environment and processed using Structure from Motion and Multi-View Stereo algorithms. This allowed the generation of reliable digital elevation models and accurate orthophotos concerning the study area. This information was subsequently used as input for mathematical models introduced to expediently and effectively identify macroplastics, improving the results obtained through visual on-site surveys. From this analysis, it was possible to deduce, following the manual detection of macroplastics in QGIS, the surface covered by plastic waste and the abundance index; these results were then used to generate thematic maps.| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi_ML.pdf
File consultabile da postazioni biblioteca
Descrizione: Elaborato finale
Dimensione
3.72 MB
Formato
Adobe PDF
|
3.72 MB | Adobe PDF |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14253/84033