杨芳, 刘君, 何南, 尹阳. 边缘互方差与互信息结合的多模医学图像配准[J]. 南昌航空大学学报(自然科学版), 2016, 30(2): 92-97. DOI: 10.3969/j.issn.1001-4926.2016.02.015
引用本文: 杨芳, 刘君, 何南, 尹阳. 边缘互方差与互信息结合的多模医学图像配准[J]. 南昌航空大学学报(自然科学版), 2016, 30(2): 92-97. DOI: 10.3969/j.issn.1001-4926.2016.02.015
YANG Fang, LIU Jun, HE Nan, YIN Yang. The Multi-mode Image Registration use Edge Correlation Deviation Combined with Mutual Information[J]. Journal of nanchang hangkong university(Natural science edition), 2016, 30(2): 92-97. DOI: 10.3969/j.issn.1001-4926.2016.02.015
Citation: YANG Fang, LIU Jun, HE Nan, YIN Yang. The Multi-mode Image Registration use Edge Correlation Deviation Combined with Mutual Information[J]. Journal of nanchang hangkong university(Natural science edition), 2016, 30(2): 92-97. DOI: 10.3969/j.issn.1001-4926.2016.02.015

边缘互方差与互信息结合的多模医学图像配准

The Multi-mode Image Registration use Edge Correlation Deviation Combined with Mutual Information

  • 摘要: 为了提高多模图像配准的精度,提出了一种改进互信息的医学图像配准方法。首先求取图像的互信息,然后通过边缘算子提取图像的边缘信息,求取图像的边缘互方差;最后,将边缘互方差和互信息通过一定的方式结合在一起,得到一个新的测度函数,并且采用遗传算法进行寻优。将本文算法用于CT图像与MRI图像进行配准,并将此配准结果和只用最大互信息的配准方法进行对比,结果表明本文算法配准结果较为准确和稳定。

     

    Abstract: In order to improve the accuracy of multimode image registration, an improved mutual information of medical image registration method is proposed. Firstly, the method calculates the mutual information of the image; then the image edge information is extracted by the edge operator, and the edge correlation deviation of image is calculated. Finally, a new measure function is got by combining the edge correlation deviation with mutual information, and the optimization solution is found by the genetic algorithm. The final result is got by the study of the registration of CT and MRI images, and the registration results are more accurate and stable than the maximum mutual information registration method.

     

/

返回文章
返回