利用相位差异技术恢复宽带白光图像

Restoration of broadband white light image using phase diversity technique

研究了用于扩展目标成像的相位差异技术。该技术利用焦面和离焦面上的两幅图像之间的相位差异来估算波前相位畸变,同时对采集的图像进行恢复。设计了白光目标成像实验系统,用宽带白光作光源,在实验室搭建实验平台,模拟无穷远处目标和波前相位畸变。采用直角梯形分光棱镜将光路分为两个通道,在一台相机上同时采集两个通道像面上的图像,消除了相机不同产生的差异,极大地抑制了噪声对图像恢复的影响,分光后系统抗两个通道独立的高斯白噪声的能力为10%。实验结果显示恢复后的图像分辨率提高了19%,表明该技术是适合光电成像系统的图像恢复技术。

The Phase Diversity （PD） technique used in wavefront detection was studied. The technique utilizes the PD between the two images on the focus and defocus to estimate the wavesfront phase aberration and to restore the degraded images. A white light imaging system was designed and a experimental platform which takes a broadband white light as a source was established, then the system above was used to simulate broadband white objects and wavefront phase aberrations in infinite distances. To grab the two images of different channels only by the same camera, a right angle trapezoid prism was designed to split the light beam into two channels, which eliminates the system deviation from different cameras and suppresses the effect of noises on the image restoration. After splitting, the denoising ability of the system for white Gaussian noise has reached 10% Moreover the experiments also demonstrate that the resolution of the restored image has increased by 19% as compared with that of a original image. These results show that the proposed method is a good mean to restore degraded images in photoelectric image systems.