利用随机中点替代法改进遥感影像分维算法

Improved Fractal Dimensional Algorithm of Romote Sensing Image with Random Midpoint Replacement Method

在总结现有分形维数计算方法的基础上,根据遥感影像的分形特性,利用随机中点替代法获取插值点,从而对目前普遍使用的三棱柱面法进行了改进。对模拟影像及实际航空遥感影像进行仿真,结果表明:改进后的三棱柱面法较好地保留了遥感影像的分形特性,计算的分形维数也更为准确。

On the basis of summarizing current fractal dimension algorithms, the random midpoint replacement algorithm was employed to obtain interpolating point according to the fractal features of remote sensing image , so the current widely used triangular prism method was improved. The simulation results with real aviation remote sensing image show that the improved method effectively keeps the image fractal features and can obtain more accurate fractal dimension.