Image Stitching Algorithm Based on Area Blocking and Optimal Seam
Qi Mu, Yang Tang · 2019
Aiming at the problem of high time complexity and poor real-time performance of image matching algorithm due to the calculation of invalid image areas and it is easy to fuse ghosts with classical fusion algorithms. An image stitching algorithm based on area blocking and optimum seam is proposed.Firstly, the overlapping areas between the matched image and the matched image are extracted quickly by using the normalized cross correlation coefficient of image energy.Then the SURF algorithm is used to extract image features and feature matching in overlapping regions.Finally, the optimal seam algorithm is used to fuse the two images to eliminate ghosts.The experimental results show that the algorithm reduces a large number of useless computations of image matching algorithm, improves the speed of image matching, and improves the quality of image mosaic. Computing time is 3.25 times faster than traditional SIFT algorithm and 1.76 times faster than traditional SURF algorithm.The proposed algorithm can realize seamless mosaic of deformed and scaled images accurately and quickly.