Ultrasound extended-field-of-view imaging based on motion estimation using quaternion wavelet

In this paper, we propose a novel extended-field-of-view (EFOV) imaging algorithm based on the property of motion estimation in quaternion wavelet domain (QWD) for ultrasound images. Firstly, transform the ultrasound video frames into QWD from spatial domain, and the result consists of one magnitude and three phases. Secondly, use two phases to estimate the motion of image sequence derived from the shift theorem. We can obtain the motion vectors through motion estimation, and the registered images through the affine transform of the motion vectors. Thirdly, use the third phase with the magnitude to fuse the registered images. Finally, the ultrasound panoramic image is achieved by means of fusing the registered images. Finally, experiments are conducted to verify the performances of the proposed algorithm.