摘要
传统量子进化算法用于搜索某些函数极值时精确度较低且稳定性较差。针对该问题,借鉴模拟退火算法,根据进化代数及个体的适应度值,修正了传统量子进化算法旋转门函数的旋转角度值,并应用于三维医学图像分割,从而形成了一种用于三维医学图像分割的改进量子进化算法。100次阈值计算实验结果表明,提出的分割算法与传统量子进化算法相比,在保持了传统量子进化算法收敛速度快特点的同时,可大大提高算法在三维分割中的精确性和稳定性。
The traditional quantum-inspired evolutionary algorithm is sometimes inaccurate and instable when it is used in searching the best solutions of a function. To solve the problem, the Simulated Annealing idea was employed into the quantum-inspired evolutionary algorithm, and formed an improved one for 3D medical images segmentation. According to the evolutionary generation and the fitness values of the solutions, the angle of the rotation gate in the traditional algorithm was modified. The thresholds searching results for 100 times show that the accuracy and the stability of the proposed algorithm that maintains the high-speed of the convergence are better than that of the traditional quantum-inspired evolutionary algorithm.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2008年第11期2942-2945,共4页
Journal of System Simulation
基金
国家博士点基金(20040699015)
关键词
医学图像
三维分割
量子进化算法
模拟退火
medical image
3D segmentation
quantum-inspired evolutionary algorithm
Simulated Annealing
