This thesis aims to use the Cityscapes dataset, a large-scale computer vision-oriented dataset, to explore possible statistical analyses applicable to this type of resource. Although originally designed for research on autonomous driving systems, Cityscapes provides valuable urban information that can be exploited to improve road safety and assess the level of livability in urban areas. The analyses conducted are based on the train set of the dataset, consisting of 2,974 annotated images collected in 18 different cities.
Questa tesi si propone di utilizzare il dataset Cityscapes, un dataset su larga scala orientato alla computer vision, per esplorare le possibili analisi statistiche applicabili a questo tipo di risorsa. Sebbene originariamente concepito per la ricerca sui sistemi di guida autonoma, Cityscapes offre preziose informazioni urbanistiche che possono essere sfruttate per migliorare la sicurezza stradale e valutare il livello di vivibilità nelle aree urbane. Le analisi condotte si basano sul train set del dataset, costituito da 2.974 immagini annotate, raccolte in 18 città diverse.
Analisi di dataset strutturati per computer vision nella pianificazione urbana
BARBATO, LUCIA
2024/2025
Abstract
This thesis aims to use the Cityscapes dataset, a large-scale computer vision-oriented dataset, to explore possible statistical analyses applicable to this type of resource. Although originally designed for research on autonomous driving systems, Cityscapes provides valuable urban information that can be exploited to improve road safety and assess the level of livability in urban areas. The analyses conducted are based on the train set of the dataset, consisting of 2,974 annotated images collected in 18 different cities.File | Dimensione | Formato | |
---|---|---|---|
Barbato_Lucia.pdf
accesso riservato
Dimensione
3.93 MB
Formato
Adobe PDF
|
3.93 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/84116