混沌PSO优化的马尔可夫随机场的深度恢复

The application of MRF based-on chaos-PSO optimization in depth information estimation

针对视觉图像的深度信息的获取,提出了一种基于马尔可夫随机场的散焦特征参数的模型,将散焦特征深度信息的获取转化为能量函数的优化问题.采用混沌粒子群优化(PSO)方法防止预测模型的局部优化,可以使预测率显著提高.实验与仿真证实了模型和算法的有效性和可能性.

For depth information estimation of vision image,a blur parameter model of defocus image based on Markov random field(MRF)was presented.It converted problem of depth estimation into optimization problem.An improved chaos-PSO algorithm was applied to complete optimization problem,which prevented that the result got into local optimization.The experiments and simulations prove that the model and algorithm are efficiency.