基于点源阵列的星载面阵CMOS相机静态PSF测量

摘要

The distribution of the optical emission from object space to image space through the camera system can be fully described by point spread function(PSF),which is important prior information for image restoration and super-resolution.In order to reduce the impact of phase and solve the problem of lack of sample points with a single point resource,the phase model is analysed and the template library based on the phase theory is established,an arithmetic is proposed to fix phase based on the template matching by similarity measure function,and then a method including target design and data processing is formed for the first time.The simulation results of the PSF measurement methods based on Monte-Carlo method show that this method is robust and the precision is up to 95％ by standard deviation of two dimension Gaussian.This method is used in the static PSF measurement of GF-4 satellite panchromatic area CMOS camera,providing references of the engineering application for the PSF measurement of the satellite borne area CMOS camera.%点扩散函数(Point Spread Function,PSF)能够完整表征物空间一点发出的光经过相机系统在像空间的分布特性,是空域图像复原、图像超分等处理的关键先验信息.针对基于点源的星载面阵CMOS相机静态点扩散函数PSF测量受点源相位影响、采样点少的问题,分析了相位模型,建立了相位理论模型模板库,提出了基于相似性度量函数模板匹配的相位确定方法,首次形成一套点源阵列靶标设计及数据处理方法,并对该PSF测量方法进行蒙特卡洛仿真,结果表明文章提出的方法对信噪比变化不敏感,具有较强稳定性,以二维高斯分布标准差衡量,测量精度达到95％.将该方法用于“高分四号”卫星全色CMOS相机的静态PSF测试,证明了该方法在星载面阵CMOS相机静态PSF测试工程应用中的可行性.