An improved PCA-SIFT algorithm application in light small UAV image registration

Unmanned aerial vehicles (UAVs) represent a quickly evolving technology, broadening the availability of remote sensing tools to small-scale research groups in a variety of scientific fields. Light small UAV remote sensing is a main way to obtain centimeter-level high resolution and hour-lever response data. Many technical skills are involved widely in the development of UAV platform covering platform development, data post-processing, and image analysis. Image registration is the key problem among them. In order to improve the performance of remote sensing images matching from Light small UAV, an improved PCA-SIFT algorithm is proposed based on the SIFT algorithm. To demonstrate the performance of the proposed method, images from DJI Phantom 4 are used for the registration experiment. The result shows that the proposed algorithm performs better in feature matching accuracy and keeps time complexity, compared with PCA-SIFT algorithm.