Machine learning guided rapid focusing with sensor-less aberration corrections
Machine learning guided rapid focusing with sensor-less aberration corrections
复制标题
机器学习引导快速对焦与无传感器像差校正
DOI：
10.1364/oe.26.030162
复制
发表时间：
2018-11-12
期刊：
影响因子：
3.8
通讯作者：
Si, Ke
中科院分区：
文献类型：
--
作者：
Jin, Yuncheng;Zhang, Yiye;Si, Ke
生物医学研究对无创、实时成像和对组织的深度聚焦有很高的要求。然而，由生物组织的折射率不均匀性引入的像差阻碍了前进的道路。本文介绍了一种基于机器学习的无传感器像差校正快速对焦方法。该方法采用卷积神经网络（Convolutional Neural Network， CNN），能快速计算出训练后具有Zernike模式的点扩散函数图像的低阶像差。结果表明，校正精度可达90%左右。每个泽尼克系数在200次重复中的平均均方误差为0.06。此外，通过在瞳孔后平面的自适应元件上加载补偿相位，可以有效地补偿1 mm厚的幻像样品和300 μ m厚的小鼠脑切片引起的像差。相位重建所需时间小于0.2 s。因此，该方法为生物科学中的活体实时成像提供了巨大的潜力。(c)根据OSA开放获取出版协议的条款，2018年美国光学学会
Non-invasive, real-time imaging and deep focus into tissue are in high demand in biomedical research. However, the aberration that is introduced by the refractive index inhomogeneity of biological tissue hinders the way forward. A rapid focusing with sensorless aberration corrections, based on machine learning, is demonstrated in this paper. The proposed method applies the Convolutional Neural Network (CNN), which can rapidly calculate the low-order aberrations from the point spread function images with Zernike modes after training. The results show that approximately 90 percent correction accuracy can be achieved. The average mean square error of each Zernike coefficient in 200 repetitions is 0.06. Furthermore, the aberration induced by 1-mm-thick phantom samples and 300-mu m-thick mouse brain slices can be efficiently compensated through loading a compensation phase on an adaptive element placed at the back-pupil plane. The phase reconstruction requires less than 0.2 s. Therefore, this method offers great potential for in vivo real-time imaging in biological science. (c) 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement