车载微光瞄准镜电子稳像技术研究

Research on Electronic Image Stabilization Technique for Vehicle-borne Low-light-level Telescope

针对"下反"稳像火控系统无法对微光夜视图像进行稳定的问题,提出了一种基于陀螺仪和块匹配相结合的实时高精度电子稳像技术。设计了电子稳像系统的总体结构;基于透视成像模型,推导了载体运动与图像运动之间的转换方程,给出了适当的图像运动模型及陀螺仪信号标定方法。在利用陀螺仪对图像运动矢量进行初始估计的基础上,结合块匹配算法对运动矢量进行精确估计;同时针对传统菱形搜索法在高噪声、低对比度微光图像中容易陷入局部极小值点的问题,基于对称校验原理对其进行改进,提高了算法的准确性。

Aimed at the stabilization problem of low-light-level image in bottom-reflecting fire control system,a novel real-time high-precision electronic image stabilization system based on gyroscope and block matching algorithm was proposed.Its overall structure was designed;the transformation equations between the carrier movement and image movement were derived on the basis of perspective imaging model;the image motion model and the gyroscope signal calibration method were given also.Based on the initial estimation for the image motion vector and combined the block matching algorithm,the motion vector was estimated accurately.The traditional diamond search method was modified to avoid local minimum in high noise and low contrast and improve the accuracy of the algorithm.