为提高基于遥感土地利用信息分类精度，本文将机载LiDAR数据和高光谱遥感数据进行融合，对基于多源遥感数据源融合的土地利用分类方法进行对比分析。结果表明，基于面向对象最邻近分类法的分类效果优于基于面向对象阈值分类法和传统监督分类法。面向对象最邻近分类结果的Kappa系数和总体分类精度分别为0.903 3和92.06%;在传统监督分类法中，最大似然分类法的分类效果最佳，其分类结果的Kappa系数和总体分类精度分别为0.81和83.5%;面向对象最邻近分类法更适用于整体的分类提取，而面向对象阈值分类法更适用于单个或少数地物种类的分类提取。

In order to improve the accuracy of remote sensing classification of land use information, this paper integrates airborne LiDAR data and hyperspectral data, and compares and analyzes land use classification methods based on multi-source remote sensing data source fusion.The results show that the classification effect of object-oriented nearest neighbor classification method is better than that of object-oriented threshold classification method and traditional supervised classification method, with Kappa coefficient and overall classification accuracy of 0.903 3 and 92.06%, respectively. Among traditional supervised classification methods, the maximum likelihood classification method has the best classification effect, with Kappa coefficient and overall classification accuracy of 0.81 and 83.5%, respectively. The object-oriented nearest neighbor classification method is more suitable for overall classification and extraction, while the object-oriented threshold classification method is more suitable for the classification and extraction of individual or a few types of ground objects.