High resolution satellite precipitation estimate using cluster ensemble cloud classification

Majid Mahrooghy, Nicolas H. Younan, Valentine G. Anantharaj, James Aanstoos

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

The link-based cluster ensemble (LCE) method is applied to a high resolution satellite precipitation estimation (HSPE) algorithm, a modified form of the Precipitation Estimation from Remotely Sensed Imagery using an Artificial Neural Network Cloud Classification (PERSIANN-CCS) algorithm. The HSPE involves the following four steps: 1) segmentation of infrared cloud images into patches; 2) cloud patch feature extraction; 3) clustering and classification of cloud patches using cluster ensemble technique; and 4) dynamic application of brightness temperature (Tb) and rain rate relationships, derived using satellite observations. The LCE method combines multiple data partitions from different clustering in order to cluster the cloud patches. The results show that using the cluster ensemble increase the performance of rainfall estimates if compared to the HSPE algorithm using Self Organizing Map (SOM). The Heidke Skill Score (HSS) is improved 5% to 7% at medium and high level of rainfall thresholds.

Original languageEnglish
Title of host publication2011 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2011 - Proceedings
Pages2645-2648
Number of pages4
DOIs
StatePublished - 2011
Event2011 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2011 - Vancouver, BC, Canada
Duration: Jul 24 2011Jul 29 2011

Publication series

NameInternational Geoscience and Remote Sensing Symposium (IGARSS)

Conference

Conference2011 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2011
Country/TerritoryCanada
CityVancouver, BC
Period07/24/1107/29/11

Keywords

  • Clustering method
  • feature extraction
  • image texture analysis
  • neural networks

Fingerprint

Dive into the research topics of 'High resolution satellite precipitation estimate using cluster ensemble cloud classification'. Together they form a unique fingerprint.

Cite this