Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA.
Phys Rev Lett. 2018 Dec 14;121(24):243902. doi: 10.1103/PhysRevLett.121.243902.
Imaging systems' performance at low light intensity is affected by shot noise, which becomes increasingly strong as the power of the light source decreases. In this Letter, we experimentally demonstrate the use of deep neural networks to recover objects illuminated with weak light and demonstrate better performance than with the classical Gerchberg-Saxton phase retrieval algorithm for equivalent signal over noise ratio. The prior contained in the training image set can be leveraged by the deep neural network to detect features with a signal over noise ratio close to one. We apply this principle to a phase retrieval problem and show successful recovery of the object's most salient features with as little as one photon per detector pixel on average in the illumination beam. We also show that the phase reconstruction is significantly improved by training the neural network with an initial estimate of the object, as opposed to training it with the raw intensity measurement.
成像系统在低光强下的性能受到散粒噪声的影响，随着光源功率的降低，散粒噪声变得越来越强。在这封信中，我们实验证明了使用深度神经网络来恢复弱光照明的物体，并展示了比经典的 Gerchberg-Saxton 相位恢复算法更好的性能，对于等效信噪比。训练图像集中包含的先验知识可以被深度神经网络利用，以检测具有接近 1 的信噪比的特征。我们将这一原理应用于相位恢复问题，并成功地恢复了物体的最显著特征，平均每个探测器像素在照明光束中只有一个光子。我们还表明，通过用物体的初始估计值而不是原始强度测量值来训练神经网络，可以显著提高相位重建。
