2020 International Conference on Computer Vision and Data Mining
Prof. Tao Cheng(程涛)

Prof. Tao Cheng(程涛)


Name:  Prof. TaoCheng

University/Department: Guangxi University of Science and Technology/School of Mechanical and Transportation Engineering

Research Area: Computational optics

Speech Title: Wide spectrum denoising (WSD) for super-resolution microscopy imaging using compressed sensing and a high-resolution camera


Because of the lack of effective denoising methods, any form of denoising is seldom performed for super-resolution microscopy, resulting in poor temporal and spatial resolutions. We propose a denoising method for STORM raw images based on compressed sensing and high-resolution cameras. This method overcomes the limitation that the raw pixel size must be approximately equal to the standard deviation of the point spread function. This method can be effectively used to remove random noise such as Poisson and Gaussian noise from very low density to ultra-high density fluorescent molecular distribution scenarios. Therefore, it is a wide spectrum denoising algorithm. Using this method, it was demonstrated that the SNR of a raw image can be increased by approximately 7 dB. Using CVX reconstruction, only 20 frames of the raw image are needed, and the time resolution is 0.86 s. The spatial resolution is also greatly improved.