Paper
1 June 2023 A road extraction method based on improved PSPNet
Xiang Jie, Rongshaung Fan
Author Affiliations +
Abstract
Aiming at the problems of low accuracy and missing of some details in road extraction by existing road extraction methods, this paper proposes a road extraction method integrating improved PSPNet. This method mainly improves the original PSPNet model by embedding SGE attention mechanism in the convolutional neural network stage to enhance the extraction accuracy. In pyramid pooling, Hybrid Dilated Convolution is used to replace global mean pooling to enhance the attention to details. In order to verify the effectiveness and feasibility of the proposed method, the improved model in this paper was compared with other common road extraction models. The experimental results show that compared with other methods, the proposed model has higher accuracy and integrity for road segmentation, and has certain practical application value.
© (2023) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xiang Jie and Rongshaung Fan "A road extraction method based on improved PSPNet", Proc. SPIE 12710, International Conference on Remote Sensing, Surveying, and Mapping (RSSM 2023), 127100G (1 June 2023); https://doi.org/10.1117/12.2682633
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Roads

Feature extraction

Convolution

Remote sensing

Education and training

Image segmentation

Data modeling

Back to Top