摘要
含有点状运动目标的红外图像情况下,本文研究了一种以最小自适应空域局部加权估计误差使全局估计误差为最小的图像背景(杂波)估计技术,给出了此类估计器性能分析和加权函数的自适应选择算法.另外,杂波抑制后,残留噪声的高斯性和独立性通过Kendall秩相关法和计算Friedman统计量的方法来进行了验证,其结果表明自适应选择的Uniform或Gabor加权系数综合性能优于同类文献中经常被单独使用的Uniform加权函数.
A new adaptive weight function selection scheme used by locally weighted regression technique for estimation of background clutter in image sequnces with moving point targets is presented in this paper. Gaussianity and independency of residuals are also verified using Kendall rank correlation and Friedman statistic methods. It is concluded that the performance of the adaptive weight function selection method presented in this paper is superior to that single Uniform weight function.
出处
《新疆大学学报(自然科学版)》
CAS
2008年第2期137-141,共5页
Journal of Xinjiang University(Natural Science Edition)
基金
国家自然科学基金资助项目(60507005
60662002)
新疆维吾尔自治区教育厅高校科研计划科学研究重点资助项目(XJEDU2005I04)