基于线性外推理论的红外焦平面阵列盲元检测算法  

Blind Pixels Detection Algorithm for IRFPA Based on Linear Extrapolation Theory

作　　者：赵春晖[1] 刘振龙 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150000

出　　处：《沈阳大学学报（自然科学版）》2013年第4期293-297,共5页Journal of Shenyang University：Natural Science

基　　金：国家自然科学基金资助项目(61077079);高等学校博士学科点科研基金资助项目(20102304110013);黑龙江省自然科学基金重点资助项目(ZD201216);哈尔滨市优秀学科带头人基金资助项目(RC2013XK009003)

摘　　要：盲元的存在严重影响了红外成像系统的性能,为有效地检测出红外焦平面阵列中存在的盲元,详细分析了盲元与点目标在邻域像素灰度分布上的差异,在此基础上利用趋势外推理论能够对事物发展预测的特性,提出了基于二维线性外推理论的盲元检测算法.实验结果证明,这一算法不仅具有复杂度低、易于实现的特点,而且具有较高的检测率,完全满足实际应用中对盲元检测精度的要求.The existence of blind pixels may cause a serious affect on the performance of infrared imaging system, in order to detect blind pixels of infrared focal plane array（IRFPA）, the differences between blind pixels and the point target of neighborhood pixel gray value distribution are analyzed in detail and the characteristics are noticed that trend extrapolation theory can predict development of things. Based on the above facts, an effective blind pixels detection algorithm is proposed based on two-dimensional linear extrapolation theory. Experimental results show that the proposed algorithm has low complexity and easy to implement, it has a higher detection rate, can fully meet the practical applications.

关 键 词：盲元 检测 灰度分布 趋势外推理论 检测率 

分 类 号：TN215[电子电信—物理电子学]