The invention discloses an unmanned aerial vehicle sequence image batch processing three-dimensional reconstruction method comprising: first, integrating the matching of images with low-precision GPS/INS information; second, creating a pole chart; third, calculating a globally consistent rotation set; forth, initializing the central point of a camera; fifth, generating a trajectory corresponding to the characteristic points; sixth, initializing a 3D structure; seventh, performing beam method difference leveling; eighth, reconstructing the dense points; and ninth, mapping the texture. According to the technical schemes of the invention, it is possible for the large field batch processing three-dimensional reconstruction of large data unmanned aerial vehicle sequence images. Through the utilization of the low-precision GPS/INS priori information to match the images, to create a pole chart, to draw the trajectory of the central point of multi-views as well as the new beam method difference leveling for function optimization, the precision and efficiency for three-dimensional reconstruction are increased.