采用CCD错位成像技术提高图像质量  

Improving Image Quality by Using CCD Subpixel Imaging

作　　者：李亚鹏[1,2] 何斌[1] 

机构地区：[1]中国科学院长春光学精密机械与物理研究所,吉林长春130033 [2]中国科学院大学,北京100049

出　　处：《光学学报》2015年第2期116-123,共8页Acta Optica Sinica

基　　金：国家863计划(863-2-5-1-13B)

摘　　要：为了克服现有电荷耦合器件(CCD)错位成像实现方法结构复杂、对移位精度要求苛刻等缺点,提出并实现了利用高精度二维平移台对单个CCD相机整体二维移位,对一定物距远处靶标成像。采用三次B样条插值、交错重组,对得到的具有既定相互位移关系的多幅相互错位半个像元的低分辨率图像进行重建,得到高分辨率图像。采用刃边法估算点扩展函数,对重建高分辨率图像进行图像复原,得到最终高分辨率图像。结果表明,采用错位成像,图像目视视觉分辨力提高了1.4倍,图像混叠效应降低,图像复原将调制传递函数面积(MTFA)由0.1577提高到0.2937。实验表明,错位成像技术能提高图像分辨力,降低图像混叠,增强图像质量,图像复原可以有效补偿重建的高分辨率图像的调制传递函数(MTF),解决了错们成像中重建图像分辨力提高但MTF下降的问题。In order to overcome the disadvantages in existing charge coupled device （CCD） subpixel imaging realization ways, such as complex structure, high subpixel precision and so on, a subpixel imaging method with single CCD camera by using high precision two-axis translation table to imaging the target at a certain distance is proposed and implemented. The multiple low-resolution images, which have half-pixel shift with each other attained by photographing target in fixed object distance using the proposed method, are interpolated and reconstructed based on cubic B spline interpolation and interlacing, and high resolution images are achieved. Then, point spread function is estimated by using knife edge method, and final high resolution image is got after performing image restoration to above reconstructed high resolution image. Results show that visual spatial resolution is improved to 1.4 times and image aliasing effect is reduced by using subpixel imaging and modulation transfer function area （MTFA） is increased to 0.2937 from 0.1577 by performing image restoration. Experiment demonstrates that subpixel imaging can enhance spatial resolution, reduce image aliasing and improve image quality. Image restoration can effectively compensate modulation transfer function （MTF） of reconstructed high resolution image, which solves the problem of spatial resolution enhancement but MTF reduction in subpixel imaging.

关 键 词：成像系统 分辨力 电荷耦合器件错位成像 图像重建与复原 调制传递函数 

分 类 号：TN386.5[电子电信—物理电子学]