ROBUST ROAD DETECTION ON HIGH-RESOLUTION REMOTE SENSING IMAGES WITH OCCLUSION BY A DUAL-DECODED UNET

被引：1

作者：

Wang, Rongfang ^{[1
]}

Wei, Haojiang ^{[1
]}

Wang, Anna ^{[2
]}

Chen, Jia-Wei ^{[1
]}

Huo, Chunlei ^{[3
]}

Niu, Yi ^{[1
]}

机构：

[1] Xidian Univ, Sch Artificial Intelligence, Xian, Peoples R China

[2] Zhejiang Expressway Construct Management Co Ltd, Hangzhou 310020, Peoples R China

[3] Chinese Acad Sci, Inst Automat, NLPR, Beijing, Peoples R China

来源：

IGARSS 2023 - 2023 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM | 2023年

基金：

中国国家自然科学基金;

关键词：

Convolution Neural Network; Dual-Decoder; Road Detection;

D O I：

10.1109/IGARSS52108.2023.10281430

中图分类号：

P [天文学、地球科学];

学科分类号：

07 ;

摘要：

It is challenging to perform robust road detection on remote sensing images in a complex scene with occlusions by plants and buildings. In this paper, an elaborate dual-decoded U-Net combined with atrous spatial pyramid pooling is proposed to tackle this scenario. In the proposed network, a dual-decoder structure is designed, where a small decoder aims to extract the attention information and it is delivered to the other decoder to enhance the context. Finally, the proposed method is verified on the DeepGlobe dataset. The experiment results demonstrate that the proposed method outperforms other compared methods.

引用

页码：5716 / 5719

页数：4