少量投影多准则CT图像重建的遗传算法研究

A Modified Genetic Algorithm of Multicriterion Image Reconstruction from Few Projections

在投影数量极少的情况下实现CT图像重建,是一不完全问题,传统FBP、ART算法一般无法解决此类问题。通过附加先验约束,引入多重准则,将其转化为多目标约束下的最优化问题,然后应用遗传算法求得全局最优解。典型算例表明,在相同条件下其重建效果要优于传统算法。

Image reconstruction from incomplete projections is an ill-posedness inverse problem which traditional FBP or ART algorithm can not solve.By adding advance constraints,it is converted into a multiple objective optimal problem,and then with a modified genetic algorithm the global optimal solution is gained at last.Typical experiments prove the results of the new algorithm are much better than the traditional's.