一种基于场景的非均匀校正算法

A Nonuniformity Correction Algorithm Based on Scene Techniques

提出了一种基于场景的非均匀校正算法,它不仅可以消除红外焦平面阵列响应特性的非均匀性,还可以补偿其漂移。建立了一个包含漂移因素的红外焦平面探测器响应模型,该模型构成了算法的基础。非均匀校正算法分两步执行:首先执行初步非均匀校正消除大部分的非均匀性,然后利用运动场景形成的红外图像序列来补偿响应特性的漂移噪声。场景运动估计是本算法的关键,块匹配方法被介绍来实现准确地运动估计。真实的红外图像数据说明了该算法的有效性。

A correction algorithm has been developed to remove nonuniformity and to compensate drift noise in the response of infrared focal-plane arrays （IRFPA） imaging system. An IRFPA sensor model, which considers IRFPA sensors＇ response drift, is set up as the basis of the algorithm. The algorithm is carried out in two steps. A preliminary nonuniformity correction is firstly carried out to remove most spatial nonuniformity of an IRFPA. Then the drift in the response of detectors is compensated with consecutive infrared image sequences, which relies on accurate estimation of infrared scene＇s motion. The formulas of the algorithm are presented, and the ability of compensating drift noise and temporal noise is evaluated. A block-matching method is also introduced to realize accurate motion estimation. Some real infrared data are finally provided to demonstrate the algorithm＇s validity.