According to the imaging characteristics of a domestic detector,the improvement and fixed-point processing are carried out for the traditional two-point NUC (Non-Uniformity Correction)and neural network NUC algorithm.After analysis of storage and data flow needs,store controller was adopted to obtain high speed access of DDR2.In situ calibration and self-adaptive iteration of NUC calibration coefficients were realized on an infrared imaging preprocessing platform based on FPGA.When the coefficients were updated,motion judgement was introduced to prevent target degradation and “ghost”.The imaging system only includes one FPGA chip,which makes the system miniaturization possible.The sufficient margin of resource enables the system to have scalable capability.Experiments prove that the method can significantly improve the uniformity of infrared imaging and suppress the time drift.
参考文献
相似文献
引证文献
引用本文
牛英宇,聂瑞杰,李丽娟.基于FPGA的红外图像非均匀校正实现方法[J].激光与红外,2016,46(8):1028~1032 NIU Ying-yu, NIE Rui-jie, LI Li-juan. Non-uniformity correction for infrared image based on FPGA[J]. LASER & INFRARED,2016,46(8):1028~1032