Abstract
Subject of study. A scheme of screening better super-resolution reconstruction based on denoising conventional raw images with wide spectrum denoising is studied. Purpose of the work. The purpose is to improve the reconstruction effect of ultra-high-resolution images. To this end, reconstruction with noise reduction and compression of ordinary raw images and high-resolution images has been investigated. Method. A binned high-resolution raw image is a conventional raw image. Conventional raw images and high-resolution raw images are denoised with wide spectrum denoising, respectively. The conventional raw images and high-resolution raw images before and after denoising are reconstructed by compressed sensing. Main Results. The denoising ability of wide spectrum denoising based on high-resolution raw images is very stable and does not change with molecular density. The signal-to-noise ratio improves by approximately 8 dB. The denoising ability of wide spectrum denoising based on conventional raw images is not good. The signal-to-noise ratios of conventional raw images is 6 dB higher than that of high-resolution raw images. The signal-to-noise ratios of denoised high-resolution and conventional raw images are almost the same. Compressed sensing reconstruction of the denoised conventional raw images is inferior to that of denoised high-resolution raw images; however, it is better than that of high-resolution and conventional raw images. Practical significance. In conventional single-molecule localizations and super-resolution microscopy, the pixel size of a raw image is about equal to the standard deviation of the point spread function. Wide spectrum denoising can improve the conventional raw image denoising and reconstruction. However, better super-resolution microscopy can be achieved based on wide spectrum denoising and high-resolution raw images. Super-resolution microscopy of high-resolution raw images will become a new research point.
© 2023 Optica Publishing Group
PDF Article
More Like This
Cited By
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Contact your librarian or system administrator
or
Login to access Optica Member Subscription