Stability analysis of hyperspectral band selection algorithms based on neighborhood rough set theory for classification

雅卡索引 高光谱成像 算法 数学 理论(学习稳定性) 粗集 维数之咒 摄动(天文学) 冗余(工程) 模式识别(心理学) 计算机科学 人工智能 数据挖掘 机器学习 操作系统 物理 量子力学
作者
Yao Liu,Junjie Yang,Yuehua Chen,Kezhu Tan,Liguo Wang,Xiaozhen Yan
出处
期刊:Chemometrics and Intelligent Laboratory Systems [Elsevier BV]
卷期号:169: 35-44 被引量:15
标识
DOI:10.1016/j.chemolab.2017.08.005
摘要

Band selection is a well-known approach for reducing the dimensionality of hyperspectral data. When the neighborhood rough set theory is used to select informative bands, different criteria of the band selection algorithms may lead to different optimal band subsets. Many studies have been analyzed the classification performance of band selection algorithms and have demonstrated that different algorithms are similar for classification. Therefore, rather than evaluating band selection algorithms using only classification accuracy, their stability should also be explored. The stability of an algorithm, which is quantified by the sensitivity of the algorithm to variations in the training set, is a topic of recent interest. Most studies on stability compare the band subsets chosen either from perturbation datasets by randomly removing methods or from perturbation datasets by cross validation methods. These methods either result in an unknown degree of overlap between perturbation datasets, or an invariable degree of overlap. In this work, we propose an adjustable degree of overlap method to construct perturbation datasets, which can set different levels of perturbation. By introducing the Jaccard index as a metric of stability, we explore the stability of six band selection algorithms based on the neighborhood rough set theory. We experimentally demonstrate that the level of perturbation, the degree of overlap, the size of the subset, and the size of the neighborhood affect stability. The results show that the maximal relevance minimal redundancy difference band selection algorithm has the greatest stability overall and better classification ability.

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
科研通AI6.4的应助被bzb采纳,获得10
刚刚
刚刚
xxxxxx完成签到,获得积分10
刚刚
略略略完成签到,获得积分10
刚刚
情怀的应助被卢西奥采纳,获得10
刚刚
丁丁完成签到 ,获得积分10
刚刚
orixero的应助被hyhy采纳,获得10
1秒前
1秒前
ouxiao277发布了新的文献求助10
1秒前
青青子衿完成签到,获得积分10
1秒前
SciGPT的应助被tian采纳,获得10
1秒前
1秒前
sxh完成签到,获得积分10
2秒前
2秒前
勾勾完成签到 ,获得积分10
2秒前
在水一方的应助被冲冲冲采纳,获得10
2秒前
大模型的应助被STAR采纳,获得10
3秒前
甜的桃子完成签到,获得积分10
3秒前
3秒前
AOM发布了新的文献求助10
3秒前
地黄饮子完成签到,获得积分10
4秒前
4秒前
威菡完成签到,获得积分10
4秒前
5秒前
33988发布了新的文献求助10
5秒前
Erik_Hsu完成签到,获得积分10
5秒前
孤狼Wwe发布了新的文献求助10
5秒前
xnhz完成签到,获得积分10
5秒前
5秒前
岳圆发布了新的文献求助30
6秒前
act完成签到,获得积分10
6秒前
刘发栋发布了新的文献求助10
7秒前
smallmoon完成签到,获得积分20
7秒前
7秒前
8秒前
wanci的应助被siuuu采纳,获得10
8秒前
搞怪的荷花完成签到,获得积分10
8秒前
9秒前
打打的应助被光亮友安采纳,获得10
9秒前
dingdingdingding完成签到,获得积分10
9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 1500
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
八旗制度の研究 600
Analyses de reconnaissances de dettes passées devant les échevins d'Ypres (1249-1291) 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 8159663
求助须知:如何正确求助?哪些是违规求助? 9555211
关于积分的说明 21725494
捐赠科研通 7749713
什么是DOI,文献DOI怎么找? 3427292
关于科研通互助平台的介绍 2552372
邀请新用户注册赠送积分活动 2471628