A high efficiency single molecule localisation algorithm with sub-pixel resolution based on fluorescence images

Abstract

We introduce a technique of fast single molecule localisation including image de-noising and Gaussian mask sub-pixel refinement (IDGM) with the characteristics of simplicity, great computational efficiency as well as high accuracy and precision. The image de-noising corrects images suffering from background and shot noise, and the Gaussian mask is used to derive the centres of single molecules with sub-pixel resolution. We quantitatively evaluate the performance of IDGM compared to two-dimensional direct Gaussian fitting (2DDGF) on 1000 simulated images at variable signal-to-noise ratios (SNRs), where the single molecule diffraction spot is simulated as a point spread function distribution. It is shown that IDGM performs better than 2DDGF in terms of accuracy and precision for all tested SNRs. In addition, the execution time is shorted by one order of magnitude, making the IDGM algorithm suitable for localising single molecules especially for on-line or time-critical applications.

Keywords:

• Single molecule localisation
• Fluorescent image de-noising
• Gaussian Mask
• PSF

Acknowledgement

We would like to acknowledge the financial support of National Science Foundation of China (NSFC) (Project No. 11104354), the Fundamental Research Funds for the Central Universities (Project No. 14CX02204A, 12CX04063A), and the Science and Technology Development Program of Qingdao (Project No. 15-9-1-92-jch, 15-9-1-19-jch).