摘要
提出了一种新的基于粒子群优化粒子滤波(PSOPF)的红外目标提取算法,将红外目标提取阈值的计算问题看作系统状态估计问题.在粒子滤波的框架下,建立了关于灰度—方差加权信息熵和像素点灰度值的阈值状态空间,建立了基于粒子群优化算法思想的系统状态转移模型,建立了基于红外目标提取效果评价函数的系统观测模型,它有效综合了红外图像中灰度、信息熵、梯度、像素点的空间位置等信息.最后,以粒子的加权平均估计目标提取的阈值.实验结果表明,该方法是有效且稳健的.
A novel infrared target extraction algorithm based on particle swarm optimization particle filter （PSOPF） was proposed. The problem of infrared target extraction was analyzed and solved in the view of state estimation. In the frame- work of particle filter, the threshold state space on the gray-variance weighted information entropy and the gray value of each pixel was established. Particle swarm optimization was introduced to construct the state transition model. The observation model based on extraction results evaluation function was constructed, which integrated gray, entropy, gradient and spatial distribution of pixels. Finally, the weighted average of all the particles was used as target extraction threshold. The experiment results prove that the proposed algorithm is effective and robust.
出处
《红外与毫米波学报》
SCIE
EI
CAS
CSCD
北大核心
2010年第1期63-68,共6页
Journal of Infrared and Millimeter Waves
基金
国家自然科学基金(60772097)
航空科学基金
关键词
粒子滤波
粒子群优化
目标提取
灰度-方差加权信息熵
particle filter
particle swarm optimization
target extraction
gray-variance weighted information entropy
