Structured Illumination Microscopy (SIM) has been widely applied in the biomedical field due to its high imaging resolution and low phototoxicity. However, SIM often requires the acquisition of 9 raw images for reconstruction, which sacrifices temporal resolution. With the emergence of Compressed Sensing (CS) technology, data compression during the sampling process has accelerated imaging speed. By combining SIM and CS techniques, it is possible to further enhance the imaging speed of SIM. In this paper, we propose an algorithm based on the aforementioned concept: utilizing the physical model of CS-SIM, we simulate the generation of a pre-training dataset. Subsequently, we train a deep learning network, namely Nonlinear Activation Free Network (NAFNet), to perform ultrafast and super-resolution reconstruction. This end-to-end approach reduces the complexity of imaging, improves imaging efficiency, and significantly enhances the quality of the reconstructed images. Now, we have achieved a compression ratio of 9:1. Furthermore, the simulated data we utilized in the training process has been generalized. Despite the limited availability of biological samples, we are still able to achieve super-resolution reconstruction of biological tissues.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.