摘要
红外焦平面阵列各探测元响应的非均匀性降低了图像的质量和温度分辨率,定标类和场景类校正算法为解决该问题提供了可行的途径.传统的基于黑体定标的校正算法由于其简单有效而被广泛使用,但是其缺点在于依赖黑体及其相关控温装置.本文提出了一种改进的基于积分时间定标的两级校正算法.在保持入射辐射不变的情况下,首先通过改变积分时间得到系列响应数据,然后利用最小二乘法估计出两点校正所需的初始校正系数,并对不同积分时间下的响应数据进行初校正,最后估计出一点校正系数.实验结果证明该方法具有计算量小、校正准确度高的优点,可取得优良的校正效果,并易于在硬件平台中实现.
The infrared image quality and temperature resolution is degraded significantly by the non- uniformity response of infrared focal pattern array. Calibration-based and scene-based non-uniformity correction algorithms provide feasible way to handle this problem. Due to its simplicity and accuracy, traditional blackbody calibration-based non-uniformity correction method is widely used but with the disadvantage of highly dependence on the existence of blackbody and its temperature control equipment to provide uniform irradiances. In this paper, an improved tw^level non-uniformity correction algorithm based on integration time calibration was proposed. Given the irradiance, the different response data is first obtained through integration time variation. Then the least square method is used to estimate the initial gain coefficients and bias coefficients for two point correction algorithm. Finally, the one point correction coefficients are estimated throutb two point correction of different response data. Experimental results show that the proposed algorithm has the merits of low complexity, high precision, excellent non uniformity correction performance, and can easily be integrated into hardware platform.
出处
《光子学报》
EI
CAS
CSCD
北大核心
2014年第1期142-146,共5页
Acta Photonica Sinica
基金
The National Natural Science Foundation of China(No.61201376)
the Western Light Project of CAS(No.Y229E21213)
关键词
非均匀性校正
积分时间
红外焦平面阵列
最
小二乘
Non-uniformity correction
Integration time
Infrared focal plane arrays
Least square
