Multiple Classifier System for Remote Sensing Image Classification: A Review

Peijun Du,Yi Liu,Sicong Liu,Junshi Xia,Wei Zhang,Kun Tan

doi:10.3390/s120404764

Abstract

Over the last two decades, multiple classifier system (MCS) or classifier ensemble has shown great potential to improve the accuracy and reliability of remote sensing image classification. Although there are lots of literatures covering the MCS approaches, there is a lack of a comprehensive literature review which presents an overall architecture of the basic principles and trends behind the design of remote sensing classifier ensemble. Therefore, in order to give a reference point for MCS approaches, this paper attempts to explicitly review the remote sensing implementations of MCS and proposes some modified approaches. The effectiveness of existing and improved algorithms are analyzed and evaluated by multi-source remotely sensed images, including high spatial resolution image (QuickBird), hyperspectral image (OMISII) and multi-spectral image (Landsat ETM+). Experimental results demonstrate that MCS can effectively improve the accuracy and stability of remote sensing image classification, and diversity measures play an active role for the combination of multiple classifiers. Furthermore, this survey provides a roadmap to guide future research, algorithm enhancement and facilitate knowledge accumulation of MCS in remote sensing community.

Highlights

Remote sensing images are widely used for land cover classification, target identification and thematic mapping from local to global scales owing to its technical advantages such as multi-resolution, wide coverage, repeatable observation and multi/hyperspectral-spectral records [1]
The accuracy assessment demonstrates that all the classifiers perform very accurately and achieve overall accuracies of 85.92% (MMDC), 92.78% (MLC), 93.41% (SVM), 93.49%
Based on the above analysis, we identified that the incorporation of diversity among the classifiers is of great importance in order to obtain better classification results via a multiple classifier system (MCS) [31]

Summary

Introduction

Remote sensing images are widely used for land cover classification, target identification and thematic mapping from local to global scales owing to its technical advantages such as multi-resolution, wide coverage, repeatable observation and multi/hyperspectral-spectral records [1]. With the advancement of remote sensing data acquisition technology, remote sensing images can be acquired by various sensors, for example, hyperspectral imaging spectrometer, high resolution sensors, polarimetric synthetic aperture radar, etc. Under this situation, image classification techniques are exposed with new challenges to process the multi-source data and serve to multidisciplinary applications [2,3]. In the past twenty years, MCS has developed rapidly and been widely used in various fields such as pattern recognition, image processing and target identification. MCS has become a hot topic in the attractive series international workshops, mainly because of its capability to improve accuracy and efficiency [11,12,13,14]

Objectives

Methods

Discussion

Conclusion