Paper
21 February 2024 Multi-level feature fusion for remote sensing image building segmentation method
Jinlei Xia, Baishou Li, Qiong Zhang
Author Affiliations +
Proceedings Volume 12988, Second International Conference on Environmental Remote Sensing and Geographic Information Technology (ERSGIT 2023); 129881A (2024) https://doi.org/10.1117/12.3024012
Event: Second International Conference on Environmental Remote Sensing and Geographic Information Technology (ERSGIT 2023), 2023, Xi’an, China
Abstract
In response to the existing issues of building edge adhesion, low extraction completeness, and precision when using high-resolution remote sensing images for building extraction, this paper proposes an improved neural network that combines residual connection modules and parallel processing of different-level resolution remote sensing images. This network effectively extracts buildings from high-resolution remote sensing images. On the WHU high-resolution remote sensing image dataset, the proposed method achieves optimal accuracy across all metrics. The precision, recall, F1 score, IoU metric, and mIoU metric of this method are 91.29%, 89.66%, 90.48%, 85.59%, and 88.06%, respectively.
(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Jinlei Xia, Baishou Li, and Qiong Zhang "Multi-level feature fusion for remote sensing image building segmentation method", Proc. SPIE 12988, Second International Conference on Environmental Remote Sensing and Geographic Information Technology (ERSGIT 2023), 129881A (21 February 2024); https://doi.org/10.1117/12.3024012
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Semantics

Feature fusion

Feature extraction

Image fusion

Convolution

Remote sensing

Image segmentation

Back to Top