| Super-resolution microscopy is a powerful tool for understanding and studying the microscopic world,and has played an important role in the development of biology,medicine,chemistry,materials and other fields.However,most existing superresolution imaging techniques rely on point-by-point scanning or computational reconstruction of multi-frame images,and their overall imaging speed is limited by factors such as scanning speed or camera sampling speed,and are only suitable for super-resolution observation of static scenes or low-speed dynamic scenes.And they are difficult to be used for fine observation of high-speed dynamic scenes.With the advancement of research on high-speed dynamic scenes in many research fields,there is an increasingly urgent need for high-speed super-resolution microscopy.To address the speed limitation of super-resolution microscopy,this thesis introduces the temporal compressive imaging method based on compressive sensing into super-resolution microscopy,and develops a series of high-speed super-resolution microscopy techniques based on compressive sensing by compressed sampling and image reconstruction of high-speed dynamic scenes,obtaining the improvement of super-resolution imaging speed under limited image sampling speed:1.Compressed wide-field fluorescence microscopy technique was developed,using compressed sampling and image reconstruction to break through the imaging speed limit of the camera.An image reconstruction algorithm based on alternating deep priories was developed to achieve high quality reconstruction of compressed images.The high-speed imaging capability of the imaging system at 2000 frames per second was verified by observation of moving fluorescent microspheres.2.Temporal compressive super-resolution microscopic imaging technique was developed to capture compressed and transient images of dynamic scenes using a dualchannel structure,and to achieve high-speed super-resolution microscopic imaging using an image reconstruction algorithm incorporating motion estimation,channel coupling and a depth network super-resolution module.The imaging speed of 1200 frames per second and the spatial resolution of 100 nm are experimentally validated.3.Compressed imaging-based structured illumination microscopy technique was proposed,in which fluorescence images under different structured illumination modes are acquired after high-speed encoding and compression,and then super-resolution dynamic scenes are reconstructed directly from the compressed images,and a superresolution imaging speed increase of nearly two orders of magnitude is obtained through computational simulation.4.Composite structured illumination temporal compressive microscopy scheme was proposed,combining composite structured illumination field and temporal compressive imaging to compress and acquire the dynamic scene to be measured,breaking through the limitation of multiple illumination in traditional structured illumination microscopy.The feasibility of the scheme is verified through computational simulations... |
