摘要
本文以仿射模型为基础将区域配准问题转化成优化问题,然后给出小波金字塔分解用于配准的一般框架,并与常规金字塔分解法进行了分析比较;针对配准过程的多变量非线性优化问题,结合混沌理论和遗传算法提出了改进的全局优化的混沌遗传算法。这种算法将混沌序列嵌入标准遗传算法,对标准遗传算法的个体进行混沌化处理,利用混沌的遍历特性来改进遗传算法的鲁棒性。
In this paper,wavelet transform is applied to solve CT/MRI medical images registration and fusion.In the process of image registration,an improved algorithm named chaos genetic algorithm is proposed,which is employed to search the global optimization solution in every each of wavelet pyramid based on rigid deformable model.The new algorithm presented in this paper combines traditional genetic algorithms with chaos theory by the full use of the property of ergodicity of chaos movement,expected to improved the performance of genetic algorithms with the help of chaos operator embedded in the algorithm-reducing the number of vain search to speed matching rate,avoiding entrapment in local minimum basin to obtain the result of higher precision.
出处
《中国医疗设备》
2008年第1期50-52,共3页
China Medical Devices
