Remote sensing data classification is an important way of information extraction and a hot research topic of remote sensing technique. Classification method of remote sensing data is an important issue, and effective selection of appropriate classifier is especially significant for improving classification accuracy. Along with the development of remote sensing technique, traditional parametric classifier is difficult to meet accuracy requirement, leading to the rapid development of intelligent algorithm based non-parametric classifiers. Recently, combined classifiers become a new hot topic for its ability of utilizing complement information of single classifier. In the present paper, characters and advantages of different classifiers as well as the research prospect are analyzed. The paper provides a scientific reference for the development of remote sensing data classification technique.