为了准确获取简单光学系统的点扩散函数(PSF), 提升图像复原质量, 本文提出了一种基于PSF测量的宽光谱PSF估计方法。首先, 测量了光学系统的窄带PSF, 并结合图像匹配算法, 标定了实际光学系统中的探测器位置和光轴中心偏移。然后, 模拟实际光学系统各波长、各视场的PSF, 再结合目标反射光谱和探测器光谱敏感信息计算实际目标的宽光谱PSF。实验结果表明: 本文提出的PSF估计方法明显优于窄带PSF估计和盲估计方法, 复原图片质量和稳定性均有明显提升, 能够准确估计实际光学成像系统的PSF。

In order to obtain point spread functions(PSFs) of a simple optical system accurately and improve the restored image quality, we present a wide-spectrum PSF estimation method based on PSF measurements. First, narrow-band PSFs are measured, and combining image matching algorithm, the sensor position and the deviation of the optical axis in the real optical system are calibrated. Then, the PSF of each wavelength and field of view is simulated and used for calculating the wide-spectrum PSFs of the real optical system according to the object reflectance spectrum and the spectral sensitivity information of the sensor. Experimental results indicate that the proposed PSF estimation method is better than the narrow-band PSF estimation and blind PSF estimation. The restored image is more stable and its quality is improved significantly. The proposed method can estimate the PSFs of the real optical imaging system accurately.