Research on cloud detection method based on voting ensemble learning using FY-4B satellite data

Jianhua Qu

doi:10.1080/10106049.2025.2487012

2025-04-07

DOI：10.1080/10106049.2025.2487012

原文链接

原文求助

摘要

En 中文

机器学习被广泛应用于卫星云检测。集成模型通常在复杂场景中比单一模型具有更优越的泛化能力。针对同质集成模型的普遍应用，本研究引入了基于FY-4B AGRI红外数据云特征训练的异构基学习器（SVM、NB、LR、DT、RF、MLP）。我们提出了投票集成学习方法，以提高稳定云检测的泛化能力和适应性。结果表明，尽管RF和MLP在单独使用时表现优异，但投票集成显著提升了整体准确率和稳定性。使用MODIS数据的交叉验证显示，该集成模型在深海、浅水、陆地和雪地四种场景下的准确率均超过91%，虚警率通常低于8%（雪地除外，为12%）。该集成模型在不同季节表现稳定。与单一模型相比，投票集成在多样化的季节和场景下展现出显著提升的适应性、准确性和鲁棒性，为复杂的云检测挑战提供了更可靠的解决方案。

Keyword:

Cloud detection

voting ensemble learning

random Forest

MLP

FY-4B satellite

期刊

Geocarto International

IF:

3.5

论文数:

2.4K

被引数:

6.9K

机构

remote sensing division, beijing huayun xingditong technology co., ltd, beijing, chinaview further author information

学者数:

论文数: 1

被引数: 0

remote sensing division, beijing huayun xingditong technology co., ltd, beijing, chinacorrespondence15737686276@163.com view further author information

学者数:

论文数: 1

被引数: 0

引用论文

A Naive Bayesian Cloud-Detection Scheme Derived from CALIPSO and Applied within PATMOS-x

Journal of Applied Meteorology and Climatology

IF0

2012-06-01

OAAI

Andrew K. Heidinger; Amato T. Evan; Michael J. Foster; Andi Walther

Assessment of Global Cloud Datasets from Satellites: Project and Database Initiated by the GEWEX Radiation Panel

BULLETIN OF THE AMERICAN METEOROLOGICAL SOCIETY

IF5.9

2013-07-01

460

OAAI

Stubenrauch, C. J.; Rossow, W. B.; Kinne, S.; Ackerman, S.; Cesana, G.; Chepfer, H.; Di Girolamo, L.; Getzewich, B.; Guignard, A.; Heidinger, A.; Maddux, B. C.; Menzel, W. P.; Minnis, P.; Pearl, C.; Platnick, S.; Poulsen, C.; Riedi, J.; Sun-Mack, S.; Walther, A.; Winker, D.; Zeng, S.; Zhao, G.

IF0

PREAI

Trends in global cloud cover in two decades of HIRS observations

JOURNAL OF CLIMATE

IF4

2005-08-01

290

OAAI

Wylie, D; Jackson, DL; Menzel, WP; Bates, JJ

IF0

PREAI

Comparison of histogram-curve fitting-based and global threshold methods for cloud detection