TY - JOUR
T1 - DENet: Double-Encoder Network With Feature Refinement and Region Adaption for Terrain Segmentation in PolSAR Images
AU - Zeng, Xuan
AU - Wang, Zhirui
AU - Sun, Xian
AU - Chang, Zhonghan
AU - Gao, Xin
N1 - Generated from Scopus record by KAUST IRTS on 2023-09-21
PY - 2022/1/1
Y1 - 2022/1/1
N2 - Recently, many studies exploit deep neural networks to promote terrain segmentation in polarimetric synthetic aperture radar (PolSAR) images. However, these works usually inherit the nature-scene approaches directly and may not be robust for the PolSAR image segmentation task. The main limitations include single-type feature construction, weak feature consistency, and geometry-agnostic collection of scattering information. In this article, we present the DENet, a double-encoder network with feature refinement and region adaption for the terrain segmentation in PolSAR images. First, a double-encoder architecture is proposed to leverage the multitype information of PolSAR images, which can provide more discriminative features than the previous methods using the single-type feature. Second, considering that the polarization information has strong consistency over the category-identical regions, a polarization-guided refinement module is proposed to maintain the feature consistency in the PolSAR segmentation model. This design alleviates the phenomenon of incomplete and fragmented segmentation results. Third, in view of the rich targets' characteristics in the scattering information, a region-adaptive convolution module is developed to facilitate the scattering information collected over the geometry-irregular regions. This design can improve the segmentation accuracy on the geometry-irregular regions. Extensive experiments are conducted on six PolSAR images to verify the effectiveness of the DENet. Compared with the previous works, our method achieves competitive performance.
AB - Recently, many studies exploit deep neural networks to promote terrain segmentation in polarimetric synthetic aperture radar (PolSAR) images. However, these works usually inherit the nature-scene approaches directly and may not be robust for the PolSAR image segmentation task. The main limitations include single-type feature construction, weak feature consistency, and geometry-agnostic collection of scattering information. In this article, we present the DENet, a double-encoder network with feature refinement and region adaption for the terrain segmentation in PolSAR images. First, a double-encoder architecture is proposed to leverage the multitype information of PolSAR images, which can provide more discriminative features than the previous methods using the single-type feature. Second, considering that the polarization information has strong consistency over the category-identical regions, a polarization-guided refinement module is proposed to maintain the feature consistency in the PolSAR segmentation model. This design alleviates the phenomenon of incomplete and fragmented segmentation results. Third, in view of the rich targets' characteristics in the scattering information, a region-adaptive convolution module is developed to facilitate the scattering information collected over the geometry-irregular regions. This design can improve the segmentation accuracy on the geometry-irregular regions. Extensive experiments are conducted on six PolSAR images to verify the effectiveness of the DENet. Compared with the previous works, our method achieves competitive performance.
UR - https://ieeexplore.ieee.org/document/9624979/
UR - http://www.scopus.com/inward/record.url?scp=85120069538&partnerID=8YFLogxK
U2 - 10.1109/TGRS.2021.3130174
DO - 10.1109/TGRS.2021.3130174
M3 - Article
SN - 1558-0644
VL - 60
JO - IEEE Transactions on Geoscience and Remote Sensing
JF - IEEE Transactions on Geoscience and Remote Sensing
ER -