About this document
Comparative Study of Loss Functions for Road Segmentation by Lamiss Kara is a document available to read on EtoBox.
This study conducts a comparative analysis of 12 loss functions for road segmentation in remote sensing imagery using the D-LinkNet model on two high-resolution datasets. The results indicate that region-based loss functions generally outperform distribution-based ones in terms of model performance metrics like F1 and IoU. The paper provides insights and recommendations for selecting appropriate loss functions tailored to road segmentation tasks.
- Author
- Lamiss Kara
- Language
- EN