基于衬底温度的红外焦平面联合非均匀性校正

分别分析了红外焦平面阵列(IRFPA)基于定标的非均匀性校正法(NUC)和基于场景的NUC算法各自的优势和问题,在此基础上提出了联合非均匀性校正方法。根据上电时刻焦平面衬底的温度值,从FLASH中提取事先存储的对应温度区间的增益和偏置校正参数,初步消除探测器的非均匀性。通过分析初步校正后图像残余非均匀性噪声的特性,提出了用具有保边缘特性的P-M滤波取代传统神经网络算法中的四邻域均值滤波来获得期望图像,从而减小了图像边缘误差。实验结果表明,该算法收敛速度快,校正精度高,有效避免了因红外焦平面响应特性漂移而引起的图像降质。

Combined nonuniformity correction algorithm of infrared focal plane arrays based on substrate temperature

The advantages and disadvantages in nonuniformity correction(NUC) algorithms based on calibration and scene of infrared focal plane arrays(IRFPA) were analysed separately. The combined NUC algorithm was presented. The thermal imaging system extracted the gain and offset factor from the FLASH which storged beforehand according to the substrate temperature of the IRFPA at the moment of power on. These factor was adopted to remove the nonuniformity of the detector simply. Based on the analysis of the characteristic of residual noise after initial correction, the P-M filter was used to replace the linear spatial average filter in the Neural Network nonuniformity correction algorithm(NN-NUC),which could preserve the image edge. Experimental results show that the proposed algorithm can accelerate the rate of convergence, reduces the correction error largely, and avoids the image degradation caused by the response drift of IRFPA.