Study on Aberration Correction of Adaptive Optics Based on Convolutional Neural Network

The existence of aberrations has always been an important limiting factor in the imaging field. Especially optical microscopy imaging, accumulated aberration system and biological samples distorts wavefront on focal plane, thereby reducing resolution. Here, we propose adaptive correction method based convolutional neural network. By establishing relationship between Zernike polynomial distorted wavefront, with help fast calculation advantage artificial intelligence network, information can be output a short time for reconstruction to achieve purpose improving Experimental results show that this effectively compensate introduced by system, agarose HeLa cells. After correcting, point spread function restored doughnut-shape, resolution cell image increased about 20%.