Study on the Land Use and Cover Classification of Zhengzhou Based on Decision Tree

Based on the data of ETM+ multi-spectral remote sensing image obtained from Landsat-7, adopting the method of decision tree-based classification, the land use and present coverage situation of Zhengzhou city are classified in this article. On the foundation of the analysis of the information content of remote sensing image bands, the sample selection in field and the laboratory test, establish the classification rules of the nodes and then formed the color region graphs of land use and vegetation coverage. Correcting the classification results and alternating above operations, the precision of classification result produced by this method is about 94 percent. The result shows that decision tree-based classification method is better than the other traditional statistical classification methods, and it can deal with noise and lost information without depending on normal school data but not need the requirement of normal distribution. It has been proved that the decision tree-based classification method has obvious advantages, such as exact classification, efficient, definite classification criterion, intuitive classification structure controllable classification precision automated classification, etc.