README

FZA_autofocus imaging is a project to achieve the autofocus in the FZA lensless imaging through the sharpness valuation WTN and get a clearer imaging by the ADMM algorithm.

Reference

• Fangyu Liu, Jiachen Wu, and Liangcai Cao, "Autofocusing of Fresnel zone aperture lensless imaging for QR code recognition," Optics Express 31, 15889-15903 (2023).
• https://doi.org/10.1364/OE.489157
• https://opg.optica.org/oe/fulltext.cfm?uri=oe-31-10-15889

WTN

• Before running this programm, confirm whether the parameters are suitable for your experiment if you have changed the input.
• The parameters you should pay attention
  • di: the distance between mask and sensor(mm)
  • dp: the size of the pixels
  • r1: the FZA parameter which is the radius of the central circle
  • theta: the weight of the gnorm in the WTN
• Running this programm, you can get the measurement image, the WTN autofocus curve and the BP reconstruction in the estimated focus distance.

ADMM

• Pay atttention to uniform the parameters between the WTN and ADMM
• After running the WTN, run the ADMM directly or record the f_distance and original image I then run the ADMM
• The parameters you can change to have different effect
  • lambda
  • mu
  • eta
  • beta
  • N
• Running this program, you can get the dynamic change display through the proess of ADMM and finally have a ADMM reconstruction image in the estimated focus distance.

Function

• FZA lensless imaging BP function
  • My...
• sharpness valuation metric
  • GNORM
  • GRA
  • LAP
  • SMD
  • ToG
  • VAR
• filtering function
  • Gauss_1

Cource

• the experiment orignal measurement image

Contact

• Fangyu Liu, liu-fy21@mails.tsinghua.edu.cn
• Jiachen Wu, wjc2022@tsinghua.edu.cn