Hybrid first and second order attention Unet for building segmentation in remote sensing images  

作　　者：Nanjun HE Leyuan FANG Antonio PLAZA 

机构地区：[1]College of Electrical and Information Engineering,Hunan University,Changsha 410082,China [2]Hyperspectral Computing Laboratory,Department of Technology of Computers and Communications,Escuela Politecnica,University of Extremadura,Extremadura E-10003,Spain

出　　处：《Science China(Information Sciences)》2020年第4期65-76,共12页中国科学（信息科学）（英文版）

基　　金：supported in part by National Natural Science Foundation of China(Grant Nos.61922029,61771192);National Natural Science Foundation of China for International Cooperation and Exchanges(Grant No.61520106001);Huxiang Young Talents Plan Project of Hunan Province(Grant No.2019RS2016)。

摘　　要：Recently,building segmentation(BS)has drawn significant attention in remote sensing applications.Convolutional neural networks(CNNs)have become the mainstream analysis approach in this field owing to their powerful representative ability.However,owing to the variation in building appearance,designing an effective CNN architecture for BS still remains a challenging task.Most of CNN-based BS methods mainly focus on deep or wide network architectures,neglecting the correlation among intermediate features.To address this problem,in this paper we propose a hybrid first and second order attention network(HFSA)that explores both the global mean and the inner-product among different channels to adaptively rescale intermediate features.As a result,the HFSA can not only make full use of first order feature statistics,but also incorporate the second order feature statistics,which leads to more representative feature.We conduct a series of comprehensive experiments on three widely used aerial building segmentation data sets and one satellite building segmentation data set.The experimental results show that our newly developed model achieves better segmentation performance over state-of-the-art models in terms of both quantitative and qualitative results.Recently, building segmentation(BS) has drawn significant attention in remote sensing applications. Convolutional neural networks(CNNs) have become the mainstream analysis approach in this field owing to their powerful representative ability. However, owing to the variation in building appearance, designing an effective CNN architecture for BS still remains a challenging task. Most of CNN-based BS methods mainly focus on deep or wide network architectures, neglecting the correlation among intermediate features.To address this problem, in this paper we propose a hybrid first and second order attention network(HFSA)that explores both the global mean and the inner-product among different channels to adaptively rescale intermediate features. As a result, the HFSA can not only make full use of first order feature statistics, but also incorporate the second order feature statistics, which leads to more representative feature. We conduct a series of comprehensive experiments on three widely used aerial building segmentation data sets and one satellite building segmentation data set. The experimental results show that our newly developed model achieves better segmentation performance over state-of-the-art models in terms of both quantitative and qualitative results.

关 键 词：BUILDING segmentation(BS) convolutional neural networks(CNNs) remote sensing high order pooling ATTENTION 

分 类 号：P237[天文地球—摄影测量与遥感] TP751[天文地球—测绘科学与技术]