基于点扩散函数模型的激光水下图像重建

Underwater Laser Imaging Restoration Based PSF Model

在激光照明条件下,如何提高水下图像的能见度问题一直是大家研究的热点。去噪和复原有助于提高成像图像的成像质量,但是,图像的分辨率仍然受到限制。图像超分辨率重建技术可以超越硬件限制而提高成像图像的分辨率,其中凸集投影法能灵活的结合先验知识,所以得到广泛的应用。先验知识如点扩散函数的精确程度能够提高重建方法的准确性。因此,本文将几种经典的点扩散函数模型,包括一个简化了的基于光线传播的点扩散模型,结合到凸集投影重建方法中,运用于一个半导体激光水下成像系统。根据实验结果,我们能得出结论:基于点扩散函数的超分辨率重建方法,能够有效地提高水下激光成像中图像重建的质量。

The visibility of underwater laser imaging has been widely used. Image restoration can help to enhance the image quality; but the resolution is still limited.Image super resolution reconstruction （SRR） can improve resolution beyond the limit of hardware, in which the Projections onto convex sets （POCS） is popular for its flexibility of incorporating prior knowledge such as the point spread function （PSF） of imaging system. A better understanding of the PSF can improve the accuracy of the SRR algrithm.The presented effort applies several PSF models including one based on laser beam propagation to the POCS method for an underwater LD-based imaging system. From experimental results, we can conclude that the combination of PSF can enhance the performance of the POCS reconstruction for underwater laser imaging.