Session
Civil Engineering, Infrastructure and Environment
Description
In this research, built-up density is quantified in Ferizaj’s urban setting through object-based classification on Planet Imagery (0.5m) acquired in 2019. The identification of homogeneous urban features is done by using ArcGIS Pro software and Mean Shift Segmentation. This segmentation is rather useful in the identification of clusters by incorporating the spatial and spectral characteristics that distinguish builtup and non-built-up regions. Training samples were obtained from satellite imagery and aerial images collected in 2019 with equal distribution of samples in the classification. Object-based classification has major benefits over pixel-based techniques in using spatial context and minimizing misclassification in urban areas. The obtained built-up density map is useful for the analysis of urban growth and the planning of sustainable land use
Keywords:
Object-Based classification, Mean Shift, Built-Up
Proceedings Editor
Edmond Hajrizi
ISBN
978-9951-982-15-3
Location
UBT Kampus, Lipjan
Start Date
25-10-2024 9:00 AM
End Date
27-10-2024 6:00 PM
DOI
10.33107/ubt-ic.2024.301
Recommended Citation
Zariqi, Pajtim and Halounova, Lena, "Using Object-Based Classification to Estimate Built-Up Density in an Urban Environment: A Case Study in Ferizaj, Kosova" (2024). UBT International Conference. 1.
https://knowledgecenter.ubt-uni.net/conference/2024UBTIC/CEIE/1
Included in
Using Object-Based Classification to Estimate Built-Up Density in an Urban Environment: A Case Study in Ferizaj, Kosova
UBT Kampus, Lipjan
In this research, built-up density is quantified in Ferizaj’s urban setting through object-based classification on Planet Imagery (0.5m) acquired in 2019. The identification of homogeneous urban features is done by using ArcGIS Pro software and Mean Shift Segmentation. This segmentation is rather useful in the identification of clusters by incorporating the spatial and spectral characteristics that distinguish builtup and non-built-up regions. Training samples were obtained from satellite imagery and aerial images collected in 2019 with equal distribution of samples in the classification. Object-based classification has major benefits over pixel-based techniques in using spatial context and minimizing misclassification in urban areas. The obtained built-up density map is useful for the analysis of urban growth and the planning of sustainable land use