ABSTRACT
ABSTRACT: An improved Speeded-Up Robust Features (SURF) algorithm is proposed to accelerate the feature points’ detection and improve the mosaic accuracy in the spine medical image mosaic. The algorithm comprises three steps: first, the Sobel operator is used to detect the edges to acquire the edge information of an image. Second, a large number of feature points are detected in the edge regions. Finally, images are spliced with the feature points obtained. Compared with the traditional SURF algorithm, the experimental results show that the precision and speed of splicing are improved by using the proposed method.
