Paper Info
Title
MP-ResNet: Multipath Residual Network for the Semantic Segmentation of High-Resolution PolSAR Images
Abstract
There are limited studies on the semantic segmentation of high-resolution polarimetric synthetic aperture radar (PolSAR) images due to the scarcity of training data and the complexity of managing speckle noise. The Gaofen contest has provided open access a high-quality PolSAR semantic segmentation dataset. Taking this opportunity, we propose a multipath residual network (MP-ResNet) architecture for the semantic segmentation of high-resolution PolSAR images. Compared to conventional U-shape encoder-decoder convolutional neural network (CNN) architectures, the MP-ResNet learns semantic context with its parallel multiscale branches, which greatly enlarges its valid receptive fields and improves the embedding of local discriminative features. In addition, MP-ResNet adopts a multilevel feature fusion design in its decoder to effectively exploit the features learned from its different branches. Comparisons with the baseline method of fully connected network (FCN with ResNet34) show that the MP-ResNet has achieved significant accuracy improvements. It also surpasses several state-of-the-art methods in terms of overall accuracy (OA), mF(1) and frequency weighted intersection over union (fwIoU), with only a limited increase of computational costs. This CNN architecture can be used as a baseline method for future studies on the semantic segmentation of PolSAR images.
Year
DOI
Venue
2022
10.1109/LGRS.2021.3079925
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
Keywords
DocType
Volume
Convolutional neural network (CNN), polarimetric synthetic aperture radar (PolSAR) image analysis, remote sensing, semantic segmentation
Journal
19
ISSN
Citations 
PageRank 
1545-598X
0
0.34
References 
Authors
0
7
Name
Order
Citations
PageRank
Lei Ding123.08
Kai Zheng200.68
Lin Dong3258.55
Yuxing Chen401.35
Bing Liu5103.24
Jiansheng Li600.68
Lorenzo Bruzzone74952387.72