An improved mosaic method based on SIFT algorithm for UAV sequence images

High-resolution images obtained by unmanned aerial vehicles (UAV) are superior in many aspects such as low cost, convenience and so on. However, the attitude of unmanned aerial vehicle cannot be achieved accurately, which leads to the decrease of accuracy and efficiency of automatic matching. There are some ascendant characteristics of SIFT operator such as invariance of scale, rotation, and lightness, and these characteristics are very useful in increasing the matching accuracy of UAV images. In this paper, features in the sequence images of UAV were matched by using SIFT algorithm, and perspective transformation and improved RANSAC algorithm were used to filter the set of matched points, and L-M algorithm was used to optimize the parameters, and at last the sequence images mosaicking were completed, and the results were satisfactory.