Title :
A Support Vector Machine-Based Particle Filter Method for Improved Flooding Classification
Author :
Insom, Patcharin ; Chunxiang Cao ; Boonsrimuang, Pisit ; Di Liu ; Saokarn, Apitach ; Yomwan, Peera ; Yunfei Xu
Author_Institution :
State Key Lab. of Remote Sensing Sci., Inst. of Remote Sensing & Digital Earth, Beijing, China
Abstract :
Support vector machines (SVMs) have been applied to land cover classification, and a number of studies have demonstrated their ability to increase classification accuracy. The high correlation between the data set and SVM training model parameters indicates the high performance of the classification model. To improve the correlation, research has focused on the integration of SVMs and other algorithms for data set selection and SVM training model parameter estimation. This letter proposes a novel method, based on a particle filter (PF), of estimating SVM training model parameters according to an observation system. By treating the SVM training function as the observation system of the PF, the new method automatically updates the SVM training model parameters to values that are more appropriate for the data set and can provide a better classification model than can the original model, wherein the parameters are set by trial and error. Various experiments were conducted using Radarsat-2 synthetic aperture radar data from the 2011 Thailand flood. The proposed method provides superior performance and a more accurate analysis compared with the standard SVM.
Keywords :
floods; image classification; land cover; learning (artificial intelligence); parameter estimation; particle filtering (numerical methods); remote sensing by radar; support vector machines; synthetic aperture radar; terrain mapping; AD 2011; Radarsat-2 synthetic aperture radar data; Thailand flood; data set selection; improved flooding classification; land cover classification accuracy; observation system; support vector machine training function; support vector machine training model parameter estimation; support vector machine-based particle filter method; Accuracy; Atmospheric measurements; Correlation; Data models; Remote sensing; Support vector machines; Training; Flooding classification; Radarsat; particle filter (PF); support vector machine (SVM);
Journal_Title :
Geoscience and Remote Sensing Letters, IEEE
DOI :
10.1109/LGRS.2015.2439575