论文标题
小脑神经网络迅速解决了涡流傅里叶编码器的反问题
Small-brain neural networks rapidly solve inverse problems with vortex Fourier encoders
论文作者
论文摘要
我们引入了一个带有列式阵列的涡旋相变，以伴随着浅，密集的``小脑''神经网络，用于高速和低光成像。我们的单发ptychographic方法利用了相干衍射，紧凑的表示和边缘增强傅立叶变形的螺旋相梯度。使用涡旋空间编码，对小脑进行了训练，以比随机编码方案实现的速度快5-20倍的图像，在噪声存在下获得更大的优势。一旦受过训练，小脑就会从仅强度数据中重建对象，求解反向映射，而无需在每个图像上进行迭代，并且没有深入学习方案。借助这种混合，光学数字，涡流傅立叶编码的小型小脑方案，我们以每秒几千帧的速度在15 W中央处理单元上以每秒数千帧的速度重建了用低光通量（5 nj/cm $^2 $）照亮的MNIST时尚对象，两个速度比卷积的Neural网络较高。
We introduce a vortex phase transform with a lenslet-array to accompany shallow, dense, ``small-brain'' neural networks for high-speed and low-light imaging. Our single-shot ptychographic approach exploits the coherent diffraction, compact representation, and edge enhancement of Fourier-tranformed spiral-phase gradients. With vortex spatial encoding, a small brain is trained to deconvolve images at rates 5-20 times faster than those achieved with random encoding schemes, where greater advantages are gained in the presence of noise. Once trained, the small brain reconstructs an object from intensity-only data, solving an inverse mapping without performing iterations on each image and without deep-learning schemes. With this hybrid, optical-digital, vortex Fourier encoded, small-brain scheme, we reconstruct MNIST Fashion objects illuminated with low-light flux (5 nJ/cm$^2$) at a rate of several thousand frames per second on a 15 W central processing unit, two orders of magnitude faster than convolutional neural networks.
