【摘要】 采用互相关代替自相关的方法计算基准图子图的相关长度,引入子图均值标准化的方法计算基准图子图的地形熵,使得它们的计算值和基准图子图的匹配概率之间具有良好的单调性.以相关长度、地形熵和粗糙度作为反映基准图子图适配性的特征向量,采用最小二乘支持向量机作为分类工具,将基准图子图划分为适配的和非适配的两类,并由适配的基准图子图类构成地形匹配区.实验结果表明所提出的方法能够有效地规划出所需的三维地形匹配区.更多还原

【Abstract】 In order to make the relationships between sub-image′s matching probability and its correlation length and terrain entropy more monotonous,cross-correlation instead of auto-correlation was employed to calculate sub-image′s correlation length and standardization of the sub-image′s mean was introduced to calculate its terrain entropy.The sub-images of a reference map were divided into two classes,one was suitable for matching and the other was not,by using the correlation length,terrain entropy and roughness to constitute the feature vector which reflecting the sub-image′s matching suitability,and by using least squares support vector machines(LS-SVM) as the classifying tool.Sub-images of the class,which was suitable for matching,make up of the suitable 3D terrain matching field.At the end,experiments were performed and the results showed that this approach was practical.更多还原