Prototype-Based Pseudo-Label Refinement for Semi-Supervised Hyperspectral Image Classification

首页 > 成果 > 详情

认领

导出

Link by DOI

反馈

作者信息关键词期刊信息基础信息归属信息摘要

成果类型：

期刊论文

作者：

Chen, Renyi;Yao, Huaxiong;Chen, Wenjing;Sun, Hao;Xie, Wei;...

通讯作者：

Sun, H;Chen, WJ

作者机构：

[Sun, Hao; Xie, Wei; Chen, Renyi; Yao, Huaxiong] Cent China Normal Univ, Sch Comp Sci, Hubei Prov Key Lab Artificial Intelligence & Smart, Wuhan 430079, Peoples R China.

[Xie, Wei; Chen, Renyi; Yao, Huaxiong; Sun, Hao] Cent China Normal Univ, Natl Language Resources Monitoring & Res Ctr Netwo, Wuhan 430079, Peoples R China.

[Chen, Wenjing] Hubei Univ Technol, Sch Comp Sci, Wuhan 430068, Peoples R China.

[Dong, Le] Xidian Univ, Sch Artificial Intelligence, Xian 710071, Peoples R China.

[Lu, Xiaoqiang] Fuzhou Univ, Coll Phys & Informat Engn, Fuzhou 350002, Peoples R China.

通讯机构：

[Chen, WJ ] H

[Sun, H ] C

Cent China Normal Univ, Sch Comp Sci, Hubei Prov Key Lab Artificial Intelligence & Smart, Wuhan 430079, Peoples R China.

Hubei Univ Technol, Sch Comp Sci, Wuhan 430068, Peoples R China.

语种：

英文

关键词：

Feature extraction;Prototypes;Training;Learning systems;Hyperspectral imaging;Sun;Geoscience and remote sensing;Class prototype;hyperspectral image (HSI) classification;pseudo-label (PL);semi-supervised learning

期刊：

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS

ISSN：

1545-598X

年：

2024

卷：

页码：

1-1

DOI：

10.1109/LGRS.2024.3385282

基金类别：

Knowledge Innovation Program of Wuhan-Shuguang Project#&#&#2023010201020377#&#&#2023010201020382 Hubei Provincial Natural Science Foundation of China#&#&#2022CFB954 National Natural Science Foundation of China#&#&#62201222#&#&#62377026

机构署名：

本校为第一且通讯机构

院系归属：

计算机学院

摘要：

Pseudo-label (PL) learning-based methods usually regard class confidence above a certain threshold for unlabeled samples as PLs, which may result in PLs still containing wrong labels. In this letter, we propose a prototype-based PL refinement (PPLR) for semi-supervised hyperspectral image (HSI) classification. The proposed PPLR filters wrong labels from PLs using class prototypes, which can improve the discrimination of the network. First, PPLR uses multihead attentions (MHAs) to extract the spectral-spatial features, and designs an adaptive threshold that can be dynamically adjusted to genera...

反馈

产权有误：本人成果被他人认领

数据有误：数据基本信息有误

归属有误：成果的院系归属、机构署名归属有误

其他原因：

验证码：

看不清楚，换一个

确定

取消

成果认领

标题：

用户	作者	通讯作者	--
	请选择	请选择	--

确定

取消

Prototype-Based Pseudo-Label Refinement for Semi-Supervised Hyperspectral Image Classification

反馈

成果认领

提示

该栏目需要登录且有访问权限才可以访问