TY - GEN
T1 - Case study on the optimal scaling factor for semantic segmentation of remote sensing image
AU - Cai, Yuanzhi
AU - Fang, Yuan
AU - Huang, Fang
AU - Fan, Lei
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - Random scaling is a widely employed training technique to mitigate overfitting in semantic segmentation models, enabling models to handle images with varying scaling factors. However, existing deep learning frameworks for remote sensing images adopt a default scaling factor of 1.0 for single-scale testing. This raises the question of whether this scaling factor is optimal for single-scale testing. To address this question, this study investigates the optimal scaling factor across six models for three remote sensing datasets. The results show that the optimal scaling factor for the UAVid, LoveDA, and Potsdam was 0.75, 1.0, and 1.25, respectively.
AB - Random scaling is a widely employed training technique to mitigate overfitting in semantic segmentation models, enabling models to handle images with varying scaling factors. However, existing deep learning frameworks for remote sensing images adopt a default scaling factor of 1.0 for single-scale testing. This raises the question of whether this scaling factor is optimal for single-scale testing. To address this question, this study investigates the optimal scaling factor across six models for three remote sensing datasets. The results show that the optimal scaling factor for the UAVid, LoveDA, and Potsdam was 0.75, 1.0, and 1.25, respectively.
KW - CNN
KW - Data augmentation
KW - Scaling factor
KW - Semantic segmentation
KW - Single-scale testing
KW - Transformer
UR - http://www.scopus.com/inward/record.url?scp=85196082328&partnerID=8YFLogxK
U2 - 10.1109/MIGARS61408.2024.10544785
DO - 10.1109/MIGARS61408.2024.10544785
M3 - Conference Proceeding
AN - SCOPUS:85196082328
T3 - 2024 International Conference on Machine Intelligence for GeoAnalytics and Remote Sensing, MIGARS 2024
BT - 2024 International Conference on Machine Intelligence for GeoAnalytics and Remote Sensing, MIGARS 2024
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2024 International Conference on Machine Intelligence for GeoAnalytics and Remote Sensing, MIGARS 2024
Y2 - 8 April 2024 through 10 April 2024
ER -