双通道分频段恒定统计非均匀性校正算法  

Dual channel constant statistic NUC algorithm of separate band

作　　者：季尔优[1] 顾国华[1] 陈钱[1] 柏连发[1] 隋修宝[1] 

机构地区：[1]南京理工大学近程高速目标探测技术国防重点学科实验室,南京理工大学光电技术系,江苏南京210094

出　　处：《激光与红外》2011年第4期474-478,共5页Laser & Infrared

基　　金：南京理工大学自主科研专项计划项目(No.2010ZDJH12)资助

摘　　要：针对基于定标的两点校正算法无法克服焦平面温漂效应和基于场景的恒定统计算法存在的鬼影现象,在对现有的算法进行了分析的基础上,提出了一种改进的恒定统计非均匀性校正算法。该算法在两点校正的基础上,对图像进行双通道分频段处理,并通过对场景运动状态的判断,有选择性地进行统计参数的更新,最后完成非均匀性校正。实验结果表明,和传统的非均匀性校正算法相比,该算法可以有效地消除红外焦平面温漂效应带来的影响,抑制基于场景的非均匀性校正算法带来的鬼影现象。An improved constant statistic NUC algorithm has been presented in this paper.This new NUC algorithm overcomes the shortages of temperature shifting effect on the FPA using the two-point correction method and the ghost phenomenon in the constant statistic correction based on the scene.This algorithm separates the raw images into different bands and two channels,selectively updates the correction parameters and completes the whole process of the NUC by means of the judgment of the scene′s movement.The experimental data shows that this algorithm can effectively eliminate the temperature shifting effect of the FPA and suppress the ghosting phenomenon caused by the NUC algorithm based on the scene.

关 键 词：非均匀性校正 恒定统计算法 鬼影抑制 运动检测 分频段处理 

分 类 号：TN216[电子电信—物理电子学] TP391[自动化与计算机技术—计算机应用技术]