Efficient Super-resolution of Phase Images Encoded with Random Phase Mask by machine learning techniques

We encoded the complex amplitude information of the measurement target into a phase image using a random phase mask (RPM) and performed super-resolution using both a convolutional neural network (CNN) and a vision transformer (ViT). This super-resolution process allowed us to encode and decode complex amplitude information into the phase image, surpassing the pixel count limitations of optical devices such as spatial light modulators (SLM) and charge-coupled devices (CCD). By applying this proposed method to virtual phase-conjugation-based optical tomography (VPC-OT), a type of optical tomography technique, we demonstrated a reduction in noise and an improvement in measurement accuracy at all depth positions of the measurement target.