With the rapid development of medical imaging technology, medical image research and application has become a research hotspot. This paper offers a solution to non-rigid registration of medical images based on ordinal feature (OF) and manifold learning. The structural features of medical images are extracted by combining ordinal features with local linear embedding (LLE) to improve the precision and speed of the registration algorithm. A physical model based on manifold learning and optimization search is constructed according to the complicated characteristics of non-rigid registration. The experimental results demonstrate the robustness and applicability of the proposed registration scheme.
As a product combining inverse synthetic aperture technology with coherent laser technology, Inverse Synthetic Aperture
Imaging Ladar (ISAIL) overcomes the diffraction limit of the telescope’s aperture, while it supplies a much better range
resolution which will not get worse at long range when the diameter telescope optics becomes smaller. Compared with
traditional microwave imaging radar, SAIL can provide a much higher-resolution image because of shorter wavelength,
and its shorter imaging time for coherent integration takes a great part in practical application. The rotational motion of
target generates Migration through Range Cells (MTRC) because of the ultra-high resolution of ISAIL. Quadratic Phase
Error (QPE) caused by Migration through Range Cells (MTRC) during the imaging time makes ISAIL image smeared. It
is difficult to estimate the QPE through traditional motion compensation algorithm. To solve this problem in the case of
uniform rotation rate, a novel QPE compensation method, based on Phase Cancellation (PC), is proposed. Firstly, a
rough range of QPE coefficient related to the wave-length, length of the target, and the rotating angle is estimated. Then,
through 1-D search, the QPE coefficient is obtained exactly. Finally, the QPE compensation is achieved. The ISAIL
imaging experiments with numerical data validate the feasibility and effectiveness of the proposed algorithm.