盲元是影响非制冷红外焦平面阵列(IRFPA)成像质量的因素之一。在对非制冷红外焦平面阵列中盲元的响应特性分析的基础上,提出一种基于双阈值快速迭代盲元检测算法。该算法基于双参考辐射源,采集辐射源两个不同温度下的成像图像,当非制冷IRFPA器件工作在线性区时,正常像元的灰度响应会随外界温度而发生变化,且该变化呈近似线性关系;此时,盲元的灰度响应几乎不随着温度的变化而改变,表现为某一固定灰度值,因此,可利用非制冷IRFPA响应的这种线性特征来检测出过热像元和死像元。实验结果验证了该算法的有效性,为非制冷IRFPA的盲元补偿提供了基础。
Blind-pixel is one of key factors that affect the imaging quality of uncooled infrared focal plane arrays(IRFPA).A particular analysis on the response characteristic of the blind-pixel was made firstly,and then a fast iterative algorithm based on double-threshold for blind-pixel detection was presented.In this algorithm,imaging pictures with two different temperatures were collected based on double reference-source respectively.When uncooled IRFPA devices operated in the linear region,the gray response of normal pixel would change with the external temperature,which showed approximatively linear relation.At the same time,the gray response of the blind-pixel was nearly unchanged with the external temperature.Then the linear characteristic of the pixel′ s response to double reference radiation source was adopted to detect the over-hot pixel and the dead pixel.The experimental results validate the effective of the presented algorithm.It provides a basis for the blind-pixel compensation for IRFPA.