为了提高非接触式测量的数据处理精度, 采用一种分阶段配准的方法, 先将缺损叶片分为4个部分, 采用自配准算法对每部分进行配准; 再对相邻两部分采用改进的完全配准算法进行整体配准。结果表明, 自配准算法与传统算法相比, 在配准误差均小于0.005 mm的前提下, 配准时间可以缩短到1 s以内; 完全配准与传统算法相比, 速度较快, 并通过0级的标准量块测量实验得出系统的测量误差小于0.010 mm, 满足叶片测量的精度要求。该分阶段配准方法对测量航空叶片具有一定的应用价值。

In order to improve the accuracy of data processing, a staged registration method was proposed. First, the defective blades were divided into four parts, and the self-registration algorithm was used to register each part; then the improved complete registration algorithm was used for the global registration of the adjacent two parts. The experimental results show that compared with the traditional algorithm, the registration time can be shorten to less than 1 s under the premise that the registration error is less than 0.005 mm by using the self-registration algorithm; the complete registration is faster than the traditional algorithm. The measurement error of the system is less than 0.010 mm by measuring the 0-level standard gauge block, which meets the accuracy precision requirements of blade measurement. The staged registration method has certain application value for the measurement of aviation blades.