Triple Contrastive Representation Learning for Hyperspectral Image Classification With Noisy Labels

计算机科学 人工智能 模式识别(心理学) 高光谱成像 特征学习 过度拟合 分类器(UML) 聚类分析 特征提取 噪音(视频) 人工神经网络 图像(数学)
作者
Xinyu Zhang,Shuyuan Yang,Zhixi Feng,Liangliang Song,Yantao Wei,Licheng Jiao
出处
期刊:IEEE Transactions on Geoscience and Remote Sensing [Institute of Electrical and Electronics Engineers]
卷期号:61: 1-16 被引量:24
标识
DOI:10.1109/tgrs.2023.3292142
摘要

Recently, Hyperspectral Image Classification (HIC) with noisy labels is attracting increasing interest. However, existing methods usually neglect to explore feature-dependent knowledge to reduce label noise, and thus perform poorly when the noise ratio is high or the clean samples are limited. In this paper, a novel Triple Contrastive Representation Learning (TCRL) framework is proposed from a deep clustering perspective for robust HIC with noisy labels. The TCRL explores the cluster-level, instance-level, and structure-level representation of HIC by defining triple learning loss. First, the strong and weak transformation are defined for hyperspectral data augmentation. Then, a simple yet effective lightweight Spectral Prior Attention-based Network (SPAN) is presented for spatial-spectral feature extraction of all augmented samples. Additionally, cluster-level and instance-level contrastive learning are performed on two projection subspaces for clustering and distinguishing samples respectively. Meanwhile, structure-level representation learning is employed to maximize the consistency of data after different projections. Taking the feature-dependent information learned by triple representation learning, our proposed end-to-end TCRL can effectively alleviate the overfitting of classifier to noisy labels. Extensive experiments have been taken on three public datasets with various noise ratios and two types of noise. The results show that the proposed TCRL could provide more robust classification results when training on noisy datasets compared with state-of-the-art methods, especially when clean samples are limited. The code will be available at https://github.com/Zhangxy1999.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
心灵美的大叔完成签到,获得积分10
刚刚
刚刚
余香的应助被绵马紫萁采纳,获得10
1秒前
2秒前
xwx完成签到,获得积分20
2秒前
汉堡包的应助被阿吟采纳,获得10
2秒前
JamesPei的应助被L123采纳,获得10
2秒前
小汪发布了新的文献求助10
2秒前
彭于彦祖的应助被1900采纳,获得30
2秒前
丘比特的应助被寒冷鸭子采纳,获得10
2秒前
李爱国的应助被科研通管家采纳,获得10
2秒前
2秒前
华仔的应助被科研通管家采纳,获得10
2秒前
天天快乐的应助被酷酷柚子采纳,获得10
2秒前
shangguang的应助被科研通管家采纳,获得10
3秒前
丘比特的应助被科研通管家采纳,获得10
3秒前
无极微光的应助被科研通管家采纳,获得20
3秒前
Maroon5发布了新的文献求助10
3秒前
勤奋一一的应助被科研通管家采纳,获得10
3秒前
刘腾发布了新的文献求助10
3秒前
慕青的应助被科研通管家采纳,获得10
3秒前
xing_xing的应助被科研通管家采纳,获得20
3秒前
Owen的应助被科研通管家采纳,获得10
3秒前
NexusExplorer的应助被科研通管家采纳,获得10
3秒前
科研通AI2S的应助被科研通管家采纳,获得10
4秒前
爆米花的应助被Xenia采纳,获得10
4秒前
4秒前
Maestro_S的应助被科研通管家采纳,获得10
4秒前
英俊的铭的应助被科研通管家采纳,获得10
4秒前
樊凡完成签到,获得积分10
4秒前
赘婿的应助被科研通管家采纳,获得10
4秒前
4秒前
LiLi完成签到,获得积分10
4秒前
4秒前
FashionBoy的应助被科研通管家采纳,获得10
4秒前
5秒前
斯文败类的应助被科研通管家采纳,获得10
5秒前
NICAI的应助被科研通管家采纳,获得10
5秒前
Maestro_S的应助被任性若枫采纳,获得10
5秒前
HOH的应助被科研通管家采纳,获得10
5秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Social Psychology in the Real World 600
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8181072
求助须知:如何正确求助?哪些是违规求助? 9571092
关于积分的说明 21816478
捐赠科研通 7777477
什么是DOI,文献DOI怎么找? 3433418
关于科研通互助平台的介绍 2557638
邀请新用户注册赠送积分活动 2479460