多聚焦岩屑图像具有点阵多，分辨率高等特点。使用基于尺度不变特征变换(SIFT) 算法或者加速鲁棒性特征(SURF) 算法在图像的自动配准上虽具有较好的配准效果，但时间效率远未达到实时性要求。本文将一种快速的图像匹配算法—定向二进制简单描述符(ORB)算法应用于多聚焦岩屑图像的自动配准。仿真实验结果表明，该方法在保证配准的精确度同时，在速度上具有较大的提升。

Multi-focus rock debris images have large numbers of pixels and high resolution. Although Scale Invariant Feature Transform(SIFT) algorithm and Speeded Up Robust Features(SURF) algorithm have met registration accuracy，their time efficiencies are both far less than the real-time requirements. A kind of fast algorithm，directional binary Oriented Brief(ORB) algorithm, is applied to multi-focus rock debris image automatic registration. Experimental results show that proposed method is efficient in registration accuracy, and its speed has been increased significantly as well.