改进的基于神经网络的红外非均匀校正算法

摘要

Aiming at the problem of traditional neural network non-uniform correction algorithm such as target degradation and slow convergence for infrared focal plane array (IRFPA), On the basis of comprehensive analysis advantages and disadvantages of traditional neural network algorithm in the non -uniformity correction, an improved non-uniform of adaptive correction algorithm based on neural network is proposed. The algorithm adopts method combined one point correction with neural network,and normalizes corresponding data,which can realize clear edge and fast convergence speed. The comparison experiment with simulated data and real IRFPA infrared data show that this scheme is effective.%目前针对红外焦平面阵列(IRFPA)传统神经网络非均匀校正算法目标退化和收敛速度慢等问题,在综合分析传统神经网路相结合算法及基础上,提出了一种改进的基于神经网络的非均匀性自适应校正算法.该算法采用一点定标与神经网络相结合的方法,并对相应数据进行归一化以实现边缘清晰和收敛速度快等目的.仿真实验以及针对实际红外图像的实验结果表明,提出的方法是合理有效的.