Improved Automatic Panoramic Image Stitching

In this book, we formulate stitching as a multi-image matching problem, and use local invariant features to find matches between all of the images because of this our method is not sensitive to ordering, orientation, scaling and illumination, its stitching precision is much higher than many other methods. We introduce an algorithm based on SIFT features to stitch panoramic images. At last, some improvements are made that we stitch the two images by using the coordinates of the image after transformation instead of interpolation; in this way complex algorithm and unnecessary calculation are avoided. Through lots of analysis and experiments we find that it is very effective.