基于亚像素配准的神经网络非均匀性校正
发布时间：2018-01-17 20:18
本文关键词：基于亚像素配准的神经网络非均匀性校正 出处：《激光与红外》2017年08期 论文类型：期刊论文
更多相关文章： 非均匀性 亚像素配准 动量项BP神经网络 收敛速度
【摘要】：红外焦平面存在严重影响成像质量的非均匀性,本文使用基于亚像素配准算法和动量项BP神经网络的非均匀性校正算法进行校正。对短波红外相机成像过程中,由于相机视轴与成像目标位置的相对偏移(由相机安装平台晃动所致),使用基于矩阵乘法的亚像素配准算法进行配准;为了加速算法收敛,采用两点法来对校正系数进行初始化;为了改善BP神经网络容易陷入局部最优值,采用增加动量项的方法来改善校正效果。通过仿真实验可以看出提出的算法消除了传统神经网络校正方法存在的鬼影和边缘模糊等问题,获得了良好的校正效果,同时提高了算法的收敛速度。为短波红外图像数据后期处理提供了良好的基础。
[Abstract]:The infrared focal plane has a serious influence on the imaging quality of non-uniformity. In this paper, the nonuniformity correction algorithm based on sub-pixel registration algorithm and momentum term BP neural network is used to correct the imaging process of short-wave infrared camera. Due to the relative deviation between the camera axis of view and the position of the imaging target (caused by the sloshing of the camera installation platform), the sub-pixel registration algorithm based on matrix multiplication is used for registration; In order to accelerate the convergence of the algorithm, the two-point method is used to initialize the correction coefficient. In order to improve the BP neural network, it is easy to fall into the local optimal value. The method of increasing momentum term is used to improve the correction effect. The simulation results show that the proposed algorithm eliminates the problems of ghost and edge blur in the traditional neural network correction method. A good correction effect is obtained, and the convergence rate of the algorithm is improved, which provides a good basis for the post-processing of short-wave infrared image data.
【作者单位】： 中国科学院上海技术物理研究所;
【基金】：全球变化与海汽相互作用专项(No.GASI-03-03-01-01)资助
【分类号】：TN215;TP391.41
【正文快照】： i引言红外焦平面探测器在航空航天遥感、消防及工业测温等领域得到广泛的应用,然而由于受材料、工艺水平等因素限制红外焦平面存在严重影响成像质量的非均R，
本文编号：1437811
本文链接：https://www.wllwen.com/kejilunwen/dianzigongchenglunwen/1437811.html
