UAV image matching based on surf feature and harris corner algorithm

The Speed-up Robust Features (SURF) algorithm has a good scale invariance in the image matching process. Its speed is fast, but it is not stable enough in the feature point extraction. Harris algorithm is an efficient corner detection algorithm, but it cannot handle the issue of scale variance in the image. Therefore, this paper considers the combination of the Speedup Robust Features algorithm and Harris algorithm in the image matching process. First, we use the Harris algorithm to extract the corner points of the two images and obtain the feature point set. Then we use the SURF algorithm to extract the feature points of the two corner set and obtain the new point set. Finally, we use the random sample consensus method to remove the error points, achieve an exact match points set and match the two images. Experiments show that the combination of the two algorithms can improve the quality of Unmanned Aerial Vehicle image matching with high efficiency and strong robustness.