The medical image registration method based on mutual information causes misregistration in monomodality image registration more readily due to the interpolation artifact giving rise to local maxima and sometimes a biased global maximum.

Thorax multimodal image registration based on artificial immune algorithm and maximization of mutual information

Based on alignment metric and normalized mutual information,image edge detection and image equalization are employed,and six registration algorithms are integrated to operate synthetic multi-modal image registrations.

Study on local extremum of object function in mutual information-based image registration;

Mono-modality image registration based on modified mutual information;

Research on CT image registration algorithm and its implementation;

Tomography image co-registration algorithm based on principle of mechanics decomposition;

Study on image co-registration based on matrix similarity degree;

In order to improve the stability and reliability of image matching,application of the scale invariant feature transform(SIFT) algorithm to image matching was studied.

The need of the image matching mode in the anti-shore engagement was expatiated on, and the shooting process was explained.

The image matching was realized by output buffer technology,the gray non-uniformity of mosaic chips was corrected by ratio average method.