Referring to the characteristics of the Micro Aerial Vehicle (MAV), a robust aerial image mosaic algorithm was presented. In this algorithm, phase correlation was used to estimate the overlapping area, in which Harris corner detector was used to extract feature points, and then a pseudo matching set was obtained. These pseudo matches were divided into inliers and outliers by using robust RANSAC algorithm. In the inliers sub-set, perspective transformation model was used to estimate the point transformation matrix between two images accurately. In the end, the image color of the overlapping band was smoothed with bilinear interpolation technique. Experimental results show that the image mosaic effect is wonderful.