摘要
提出了一种基于自适应免疫遗传算法的求解最小权三角划分(MWT)问题的方案,通过自适应地调整疫苗库的进化和有选择地注射疫苗,提高了新算法的收敛速度和全局搜索能力,结合具体的MWT问题,给出了疫苗更新与注射算子构造的具体方案。仿真实验表明,新算法能产生比免疫算法更好的划分效果,尤其适合大规模点集,有较大的实用价值。
The problem of minimum weight triangulation （MWT） is one of the most important issues in computer vision. This paper proposes an adaptive immune genetic algorithm（AlGA） to solve the problem. Based on the analysis of immune algorithm（IA） properties, the convergence speed of AlGA is faster than IA and the global search capability is improved with a self-adaptive adjustment method to the vaccine pool together with selected vaccination. According to the practical MWT, the strategies of updating and injecting a vaccine for the problem are both provided in the paper. Simulation results show that the algorithm performs better than IA in terms of quality of MWT, especially for the large scale of point cluster, and has good practical value.
出处
《计算机工程》
CAS
CSCD
北大核心
2007年第21期189-191,共3页
Computer Engineering
基金
国家自然科学基金资助项目(69775022)
国家"863"计划基金资助项目(863-306-ZT04-06-3)
关键词
最小权三角划分
免疫算法
疫苗
计算机视觉
minimum weight triangulation（MWT）
immune algorithm（IA）
vaccine
computer vision
