Central Attention Network for Hyperspectral Imagery Classification

被引：109

作者：

Liu, Huan ^{[1
]}

Li, Wei ^{[1
]}

Xia, Xiang-Gen ^{[1
,2
]}

Zhang, Mengmeng ^{[1
]}

Gao, Chen-Zhong ^{[1
]}

Tao, Ran ^{[1
]}

机构：

[1] Beijing Inst Technol, Sch Informat & Elect, Beijing 100081, Peoples R China

[2] Univ Delaware, Dept Elect & Comp Engn, Newark, DE 19716 USA

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS | 2023年 / 34卷 / 11期

基金：

中国国家自然科学基金; 北京市自然科学基金;

关键词：

Feature extraction; Data mining; Convolutional neural networks; Atomic measurements; Hyperspectral imaging; Transformers; Spectral analysis; Central attention; hyperspectral imagery (HSI); spectral-spatial feature extraction; transformer; DIMENSIONALITY REDUCTION; CNN;

D O I：

10.1109/TNNLS.2022.3155114

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this article, the intrinsic properties of hyperspectral imagery (HSI) are analyzed, and two principles for spectral-spatial feature extraction of HSI are built, including the foundation of pixel-level HSI classification and the definition of spatial information. Based on the two principles, scaled dot-product central attention (SDPCA) tailored for HSI is designed to extract spectral-spatial information from a central pixel (i.e., a query pixel to be classified) and pixels that are similar to the central pixel on an HSI patch. Then, employed with the HSI-tailored SDPCA module, a central attention network (CAN) is proposed by combining HSI-tailored dense connections of the features of the hidden layers and the spectral information of the query pixel. MiniCAN as a simplified version of CAN is also investigated. Superior classification performance of CAN and miniCAN on three datasets of different scenarios demonstrates their effectiveness and benefits compared with state-of-the-art methods.

引用

页码：8989 / 9003

页数：15

共 42 条

[1] Classification of hyperspectral data from urban areas based on extended morphological profiles
Benediktsson, JA
Palmason, JA
Sveinsson, JR
[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2005, 43 (03): : 480 - 491
[2] Learning With l1-Graph for Image Analysis
Cheng, Bin
Yang, Jianchao
Yan, Shuicheng
Fu, Yun
Huang, Thomas S.
[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING, 2010, 19 (04) : 858 - 866
[3] Morphological Attribute Profiles for the Analysis of Very High Resolution Images
Dalla Mura, Mauro
Benediktsson, Jon Atli
Waske, Bjoern
Bruzzone, Lorenzo
[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2010, 48 (10): : 3747 - 3762
[4] Devlin J, 2019, 2019 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES (NAACL HLT 2019), VOL. 1, P4171
[5] Dosovitskiy A., 2020, INT C LEARN REPR
[6] Local Constraint-Based Sparse Manifold Hypergraph Learning for Dimensionality Reduction of Hyperspectral Image
Duan, Yule
Huang, Hong
Tang, Yuxiao
[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2021, 59 (01): : 613 - 628
[7] Local Manifold-Based Sparse Discriminant Learning for Feature Extraction of Hyperspectral Image
Duan, Yule
Huang, Hong
Li, Zhengying
Tang, Yuxiao
[J]. IEEE TRANSACTIONS ON CYBERNETICS, 2021, 51 (08) : 4021 - 4034
[8] Fukunaga K., 2013, INTRO STAT PATTERN R
[9] A CNN With Multiscale Convolution and Diversified Metric for Hyperspectral Image Classification
Gong, Zhiqiang
Zhong, Ping
Yu, Yang
Hu, Weidong
Li, Shutao
[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2019, 57 (06): : 3599 - 3618
[10] HSI-BERT: Hyperspectral Image Classification Using the Bidirectional Encoder Representation From Transformers
He, Ji
Zhao, Lina
Yang, Hongwei
Zhang, Mengmeng
Li, Wei
[J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2020, 58 (01): : 165 - 178

← 1 2 3 4 5 →