.png){kind=icon}
.png){kind=icon}
Return
Wavelet-domain autofocusing algorithm for lensless ptychographic imaging\
.png){kind=icon}
PRE
AI
DOI：10.1016/j.measurement.2025.117634.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Misalignment of the specimen-detector distance introduces perturbations in lensless ptychographic scanning coordinates, leading to reconstruction artifacts caused by discrepancies between the inversion algorithm and experimental uncertainties. Here, we propose an accurate and efficient autofocusing algorithm to calibrate the specimen-detector axial distance in lensless ptychography imaging online. The proposed algorithm leverages wavelet-domain multidimensional coefficients to quantify reconstruction clarities in virtual planes within the depth of field, which requires neither careful window searching nor parameter fine-tuning during the iteration process and can effectively escape from local optimums to avoid violent convergence oscillation and crosstalk. Simulations and experiments conducted on both amplitude and biological specimens demonstrate that the proposed algorithm achieves robust convergence in nearly hundred iterations, significantly removes reconstruction artifacts, and delivers a several-fold to orders of magnitude improvement in convergence speed, calibration accuracy and convergence uncertainty compared to conventional sharpness-based autofocusing. These advancements substantially broaden the potential applications, including other coherent diffractive imaging such as coded ptychography (coherent modulated imaging), in-line holography, and multi-plane phase retrieval.
Keywords:
Autofocusing algorithm
Wavelet transform
Ptychography
Calibration accuracy
Calibration uncertainty