AdaptiveOptics.simulation

Simulation for sensorless adaptive optics (Confocal microscopy, Modal method)

Introduction

The simulation generates the aberrated PSF on the pupil with Zernike polynomial 5-22 orders, and corrects them with different methods:

1. GA (Genetic Algorithm). 1000+measurements
2. Modal method. 350+measurements
3. SPGD (Stochastic Parallel Gradient Descent). 250+measurements
4. Optimal Modal method. 150+measurements

All the methods follow the basic direction---> Estimate the Zernike polynomial coefficients based on images with some of measurements.

<p> <img src='img/Comparison of different metrics.png' align="right" width=400> </p>

A comparison of different metrics (the steeper the curve, the better the metric).

• M1: Variance (aberration.m);
• M2: Sharpness (aberration2.m);
• M3: Gradient (aberration3.m);
• M4: Summation (aberration4.m). <br><br><br><br><br><br><br><br><br>

Attention

This repo. is made just for self using. It may have bugs or trouble. You can contact me with a github issue.

If you find it useful, please cite our work:

Liu, J., Zhao, W., Liu, C., Kong, C., Zhao, Y., Ding, X., & Tan, J. (2019). Accurate aberration correction in confocal microscopy based on modal sensorless method. Review of Scientific Instruments, 90(5), 053703.