【摘要】 提出一种对参数活动轮廓模型中能量系数和轮廓线进行同步优化的方法,并初步探讨对能量系数学习问题。该方法借鉴免疫优化思想,以能量系数和轮廓线构成的四维向量为解,以控制点移动次数和目标分割误差为评价指标,通过免疫克隆选择算法和贪婪算法的有机融合,实现能量系数和轮廓线的同步优化。仿真实验结果表明,该方法是有效的。更多还原

【Abstract】 A simultaneity optimization algorithm based on Immune Clone Selection(ICS) is proposed for searching optimal energy coefficient and contour of Parameter Active Contour Model(PACM).In the optimization problem,the solution is defined as a four dimensional vector which is composed of energy coefficients and contour,the objective function is defined as the combination of segmentation error rate and contour-point-shift mean.By testing to some representative images,it finds the algorithm is effective.更多还原