Multiple-view change caused by small unmanned aerial vehicles (UAVs) monitoring the ground, result in image distortion, multi-view transformation and low overlap. Thus, such change has a strong effect on the accuracy of image registration. In this study, we utilize a Siamese network to deal with the complexity registration of low-altitude remote-sensing images. A robust neighbor-guided patch representation is designed to describe feature points based on neighborhood relation reconstruction and patch selection. The network is trained based on rotation-invariant layer to solve the inevitable rotation and nonrigid deformation caused by multi-view images in low-altitude remote-sensing images. With only 3 training images involving 4,500 putative matches, the experiment results demonstrated that the learned network can process the scenarios of yaw rotation, pitch rotation, mixture, and extreme(e.g. mixture, scaling and distortion occur simultaneously) of UAV better than other six state-of-the-art methods.