RADet: Refine Feature Pyramid Network and Multi-Layer Attention Network for Arbitrary-Oriented Object Detection of Remote Sensing Images

被引:109
作者
Li, Yangyang [1 ]
Huang, Qin [1 ]
Pei, Xuan [1 ]
Jiao, Licheng [1 ]
Shang, Ronghua [1 ]
机构
[1] Xidian Univ, Sch Artificial Intelligence,Minist Educ, Int Res Ctr Intelligent Percept & Computat,Key La, Joint Int Res Lab Intelligent Percept & Computat, Xian 710071, Peoples R China
基金
中国国家自然科学基金;
关键词
remote sensing; arbitrary-oriented object detection; feature pyramid network; attention mechanism; mask; SEGMENTATION;
D O I
10.3390/rs12030389
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Object detection has made significant progress in many real-world scenes. Despite this remarkable progress, the common use case of detection in remote sensing images remains challenging even for leading object detectors, due to the complex background, objects with arbitrary orientation, and large difference in scale of objects. In this paper, we propose a novel rotation detector for remote sensing images, mainly inspired by Mask R-CNN, namely RADet. RADet can obtain the rotation bounding box of objects with shape mask predicted by the mask branch, which is a novel, simple and effective way to get the rotation bounding box of objects. Specifically, a refine feature pyramid network is devised with an improved building block constructing top-down feature maps, to solve the problem of large difference in scales. Meanwhile, the position attention network and the channel attention network are jointly explored by modeling the spatial position dependence between global pixels and highlighting the object feature, for detecting small object surrounded by complex background. Extensive experiments on two remote sensing public datasets, DOTA and NWPUVHR -10, show our method to outperform existing leading object detectors in remote sensing field.
引用
收藏
页数:20
相关论文
共 36 条
[1]  
[Anonymous], 2018, COMPUTER VISION PATT
[2]  
[Anonymous], 2017, ARXIV170609579
[3]  
[Anonymous], 2017, PROC IEEE C COMPUT V
[4]  
[Anonymous], PROC CVPR IEEE
[5]  
[Anonymous], 2017, IEEE I CONF COMP VIS, DOI DOI 10.1109/ICCV.2017.322
[6]  
[Anonymous], IEEE T PATTERN ANAL
[7]  
[Anonymous], PROC CVPR IEEE
[8]  
[Anonymous], ADV NEURAL INFORM PR, DOI DOI 10.1109/TPAMI.2016.2577031
[9]  
[Anonymous], 2017, P 31 INT C NEURAL IN
[10]  
[Anonymous], 2014, Neural Information Processing Systems