物体表面重建轮廓拼接的改进遗传算法

Contour Tiling of Object Surface Reconstruction Based on a Modified Genetic Algorithm

为解决物体表面重建中的轮廓拼接问题,将其转化为在有向图中寻求最优路径问题.提出了基于遗传算法的适用各种目标函数的轮廓拼接算法,其中对初始种群的产生、交叉算子和变异算子等做了改进,以确保产生的个体均能代表有效解.算例模拟结果表明,该算法简单可行,在优化性能、收敛速度及鲁棒性等方面优于模拟退火算法.

The contour tiling problem in object surface reconstruction was transformed to a problem of searching for an optimal path in a directed graph. An improved generic algorithm （GA） was proposed to solve the objective functions related to the contour tiling problem. The improvements on initial population, crossover and mutation were made to the conventional GA to ensure any individual to present a valid resolution. The result of an example shows that the proposed algorithm is simple and feasible, with better performances in optimization, convergence speed and robustness compared with simulated annealing for contour tiling.