为分析最大窗口面积和多种分辨率数字高程模型（digital elevation model, DEM）对地形起伏度提取最佳面积的影响，探究定量分析最佳面积的可行方案，并依据多种类型数据最佳结果进行地貌划分。以获取的先进对地观测卫星（advanced land observing satellite, ALOS）、先进星载热发射和反射辐射仪全球数字高程模型（advanced spaceborne thermal emission and reflection radiometer global digital elevation model, ASTER GDEM）的第2版本和航天飞机雷达地形测绘任务（shuttle radar topography mission, SRTM）的3角秒精度共3种DEM数据为基础，通过调用Arcpy模块，利用均值变点法，分别以最大窗口面积和DEM分辨率作为单一变量，统计分析影响机理。为确定适合的最大窗口面积，提出了使用遍历方式对起伏度分类结果与DEM数据进行相关性分析，根据最大相关系数，确定不同数据的最佳面积，并套合海拔分类数据得到地貌分布图。结果表明：最佳面积随着最大窗口面积的增大而呈阶梯式增加；在同一面积范围内，DEM分辨率与最佳面积存在负相关关系，即随着分辨率增大，对应面积按序减小；依据相关分析，得到ALOS DEM的最大相关系数为0．785，对应的最佳面积为0． 21 km²；ASTER GDEM． V2 DEM为0．764，面积为0．32 km²；SRTM3 DEM为0．782，面积为0． 40 km²；得到3种数据最佳面积相应的地貌分类图，通过与中国1∶ 100万地貌类型空间分布数据图比较，发现实验数据地貌分布较为合理，层次较为丰富，边界较为清晰。可见最大窗口面积相比于DEM分辨率对最佳面积影响更为显著，为解决最大窗口面积影响，相关分析可以为定量确定地形起伏度最佳面积提供理论依据。

In order to analyze the effect of maximum window area and multi-resolution DEM (digital elevation model) on the extraction of optimal area for relief amplitude, explore feasible solutions to quantitatively analyze the optimal area, and classify landforms based on the best results of various types data. Based on three DEM data obtained, including ALOS (advanced land observing satellite), ASTER GDEM. V2 (Version 2 of the advanced spaceborne thermal emission and reflection radiometer global digital elevation model) and SRTM3 (3 arc-seconds shuttle radar topography mission), the influence mechanism was statistically analyzed by calling the Arcpy module and using the mean variation point method, respectively, with the maximum window area and the resolution of the DEM as a single variable. In order to determine the appropriate maximum window area, the traversal method to correlate the relief amplitude classification results with the DEM data were proposed, to determine the optimal area for different data according to the maximum correlation coefficient, and to arbitrate the elevation classification data to obtain the geomorphic distribution map. The results show as follows. The optimal area increases in a stepwise manner with the increase of the maximum window area. There is a negative correlation between the DEM resolution and the optimal area in the same area range, that is, the corresponding area decreases sequentially with the increase of resolution. Based on the correlation analysis, the maximum correlation coefficient of ALOS DEM is 0.785 0, which corresponds to the optimal area of 0.21 km². ASTER GDEM. V2 DEM is 0.764, with an area of 0.32 km², and SRTM3 DEM is 0.782, with an area of 0.40 km². The geomorphological classification maps corresponding to the optimal areas of the three data were obtained, and by comparing them with the spatial distribution of China's 1∶1 million geomorphological types, it is concluded that the distribution of geomorphological features of the experimental data is more reasonable, and the boundaries are more clear. It is concluded that the maximum window area has a greater influence on the optimal area compared to the DEM resolution, and in order to solve the influence of the maximum window area, the correlation analysis can provide a theoretical basis for quantitatively determining the optimal area of relief amplitude.