Multi-channel descriptors and ensemble of Extreme Learning Machines for classification of remote sensing images
作者：
Highlights：
• An accurate method for real-time remote image classification.
• It avoids the time-consuming descriptor learning step as well as kernelized classifiers.
• The method combines multi-channel Gabor-based descriptors and an ensemble of ELM.
• The developed Gabor-based descriptor outperforms conventional grayscale BGP.
• Outperfoms Linear SVM and V-ELM, and has lower complexity than kernelized SVM.
摘要
•An accurate method for real-time remote image classification.•It avoids the time-consuming descriptor learning step as well as kernelized classifiers.•The method combines multi-channel Gabor-based descriptors and an ensemble of ELM.•The developed Gabor-based descriptor outperforms conventional grayscale BGP.•Outperfoms Linear SVM and V-ELM, and has lower complexity than kernelized SVM.
论文关键词：Scene classification,Multi-channel descriptor,Extreme learning machine,Ensemble of classifiers
论文评审过程：Received 7 April 2015, Revised 4 September 2015, Accepted 8 September 2015, Available online 16 September 2015, Version of Record 1 October 2015.
论文官网地址：https://doi.org/10.1016/j.image.2015.09.004
