A novel algorithm for multimodality medical image registration based on IPSO algorithm and hybrid information
Yizhong Song · Journal of Optoelectronics·laser · 2009
There are some problems such as registration inaccuracy and slow convergence in image registration when using mutual information of images as similarity function.Considering this problems,firstly,the measure function of hybrid information is presented by combining improved morphology gradient with mutual information,which is based on the pixels and structure information of the images.Secondly,the improved particle swarm optimization(IPSO) algorithm is adopted to search the optimum parameters in the medical image registration.Lastly,a new medical image registration method is presented by using the IPSO and hybrid information.The experimental result shows that the new registration algorithm can increase the registration′s robustness and easily achieve accurate registration result,and it can be used to meet the need of the multimodality medical image registration.