计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (22): 31-33.

• 学术探讨 • 上一篇    下一篇

基于QPSO的纹理合成算法

周 焘,须文波,孙 俊   

  1. 江南大学 信息工程学院,江苏 无锡 214000
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-08-01 发布日期:2007-08-01
  • 通讯作者: 周 焘

Texture synthesis algorithm utilizing Quantum-behaved Particle Swarm Optimization

ZHOU Tao,XU Wen-bo,SUN Jun   

  1. Department of Information,Jiangnan University,Wuxi,Jiangsu 214000,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-08-01 Published:2007-08-01
  • Contact: ZHOU Tao

摘要: 针对PSO算法搜索空间有限,容易陷入局部最优点的缺陷,提出一种以块算法为基础,量子粒子群优化算法(QPSO)为优化策略的纹理合成方法。实验结果表明,与标准PSO算法相比,由于量子粒子群优化算法(QPSO)显著的全局收敛性,这种新型的纹理合成方法,使最后的合成图像中采样块结合处更流畅,纹理更细腻。

关键词: 纹理合成, 块采样, 粒子群优化算法(PSO), 量子化粒子群算法(QPSO)

Abstract: Aimed at solving the problem of limited searching space and local optima shortcoming,a new patch-based texture synthesis method utilizing Quantum-behaved Particle Swarm Optimization(QPSO) is proposed.Compares with basic PSO algorithm,this texture synthesis method can improve the exquisiteness of the texture and the joints can become more fluent just because of the evident global convergence of QPSO.

Key words: texture synthesis, patch-behaved sampling, Particle Swarm Optimization(PSO), Quantum-behaved Particle Swarm Optimization(QPSO)